- Learning center
- Google Analytics
1.1 Introducing Google Analytics
In the early days of the internet, it might have made some sense just to establish a web presence without worrying too much about how well it worked. Simply being there was the vital thing. These days, however, every business is on the web, from the biggest corporation to the tiniest start-up, and the competition for clicks is fierce. You can’t rely on luck to attract visitors, let alone convert them into buyers. That’s why you invest in SEO, social media and other modern forms of marketing. But are you taking the next step and analyzing the effectiveness of all these? If you’re not, then you’re still just guessing – still just throwing it out there and relying on luck.
That’s why you need Google Analytics: to measure and analyze the effectiveness of your sites and apps, along with the marketing associated with them.
What is Google Analytics?
Google Analytics is a service from Google that enables you to track and analyze a wide range of activity across a range of online platforms: your website, iOS/Android app and even internet-connected devices such as point-of-sale systems and games consoles. To begin, you simply sign up and enter the details of the website or app. The system generates a tracking code, which you need to add to the code of your site or app. (There’s a custom Measurement Protocol that needs to be implemented on other platforms and devices, but the principle is the same.) The tracking code collects usage data from your site or app and sends it to Google Analytics, where it’s processed and made available in a range of versatile reports, which you access online. 
The scope of the reports in Google Analytics is breathtaking.  Not only can you can find out how many users or visitors you’ve had, but what browser and OS they used, where they were located, what their interests are (based on their previous browsing history) and what brought them to your site: paid or organic search, referral from another site or a social network, an email campaign or newsletter, and so on. You can track exactly what they did on your site – which pages they arrived at, which pages they visited next and – with the implementation of some optional custom code in your tracking code – what interactions they had with elements within a page like Flash content, Ajax or embedded video. For e-commerce sites, you can track a customer’s entire path from initial visit through to checkout – or if they drop out along the way, you can see exactly where, so you can identify any trouble spots in your purchase funnel.
That’s the point, really: Google Analytics gives you the information you need to analyze your site or app, identify any weak spots and take informed action to improve its performance. On the marketing side, in addition to audience characteristics and behavior, Analytics can also report on which marketing campaigns and channels are driving the most traffic to your site, and which are eventually delivering the most successful outcomes. These are determined by the goals you define for the site, including such objectives as lead generation, newsletter sign-ups, brand building and so on, in addition to purchases. It means you can make an informed decision on where to direct your media spend most effectively.
Google Analytics can also assist you with search engine optimization to attract more new visitors and remarketing campaigns to entice previous visitors to return and complete a purchase or purchase again. Analytics can even make it simple to conduct content experiments to test whether changes to your site produce better outcomes: you simply set up variants of a page and Analytics takes care of the rest, redirecting visitors randomly to each variant, tracking what happens next and presenting the results in easily-digested reports so you can make your decision.
Sounds amazing, right? It doesn’t end there: Google Analytics enables you to configure your data and customize reports to suit your specific business needs, to share information and collaborate with colleagues, and integrate your Analytics account with other services such as Google AdWords and AdSense so you can analyze the results of your marketing activities in depth and get the best from your ad spend. Even more amazingly, all this is free of charge (although there is a Premium version offering additional options). 
Web analytics trends and statistics
With all this on offer, it’s perhaps surprising that not everyone is using Google Analytics. According to an early 2014 survey of American marketing professionals, only 32.5% of projects use any kind of marketing analytics at all.  Against this, however, W3Techs conducted a web technology survey of the top ten million websites as determined by popularity rankings provided by Alexa (an Amazon.com company). It used an average ranking over three months and found that Google Analytics is used by exactly 50.0% of all these websites, which equates to a traffic analysis tool market share of 81.6%. 
In a mid-2014 survey of business executives, managers and analysts from around the world, 66% of respondents claimed to have gained a competitive advantage from their use of analytics.  It must be noted that often these professionals have undertaken training from a Google Analytics workshop. In another survey, marketing professionals (mainly senior ones) were asked how important new marketing technologies were to their group’s overall effectiveness and performance: 29% described them as “essential” and 38% “very important”. When asked why, 66% of respondents said their investment in such technologies enabled them to achieve a more targeted, efficient and relevant customer, and 54% a greater return and accountability of marketing/advertising spend. In addition, 29% claimed increased productivity, while 26% cited improved rates of conversion, closure and deal value.
The greatest challenges they identified were the integration and centralization of increasingly fragmented data (54%) and figuring out what to select and how to integrate (48%).  It’s not too much of a stretch to speculate that the power of Google Analytics to pull together and organize disparate types of data could assist you to achieve some of the impressive benefits mentioned.
Looking ahead, a wide-ranging international survey in mid-2014 – covering marketers, advertisers, service providers, technologists and publishers across 17 markets – found broad agreement across those markets that the value of data-driven marketing and advertising (DDMA) is growing, and almost three-quarters of respondents expected to increase their spending on DDMA in the following year. Correspondingly, two areas were identified as most likely to experience the greatest rise in spending: digital campaign execution, very closely followed by audience analytics, measurement and attribution. 
Can we use Google Analytics to measure return on investment?
Audience analytics is on a rising trend, therefore, and Google Analytics is by far the market-leading choice for the job. Managers and marketing professionals believe it has the potential to deliver the right customers and improve the focus of their marketing and advertising spend. The detailed data Google Analytics provides can help you make informed decisions to improve the performance of your site or app through better-targeted content, user flow improvement and conversion rate optimization. It can help you to assess the whole range of your marketing activities, including SEO, SEM and online display advertising, so you can direct your spend where it does the most good and hence increase the return on your investment.
The costs of implementing and using Google Analytics can be very modest – anything from a few hours of your time for “DIY Analytics” to the ongoing cost of hiring an Analytics expert – but the return on this investment could be a significant boost to your business. Putting a figure on it could involve a lot of complicated and subtle calculations – one expert has devised a complex formula “full of specific computations of revenue incrementally delivered for various analytical efforts”  – but your actual gain will depend to a large extent on how wisely you use the information Analytics gives you. It boils down to this: to make informed business decisions, you need solid data. Let’s take a look at what Google Analytics can deliver for you and your brand or business.
1.2 What is the business case for using Google Analytics
What can Google Analytics do for you? Let’s take a look at the potential benefits it offers for your business.
Understanding the business opportunity
The key to informed planning and decision-making is having reliable data. There are plenty of ways to do simple things like count clicks or visitors, but Google Analytics enables you to measure many more things in much more detail and answer much more sophisticated questions. If it seems complex and daunting to start with, that’s because Google Analytics offers you so many options. Once you’ve an idea of the questions you’re interested in, Google Analytics will almost certainly help you find the answers.
For example, you might be assuming that new visitors are finding your site by typing in its URL or searching for the company name. But is that in fact the case? Google Analytics can tell you exactly what search terms have led people to your site, and give you an idea of which keywords should be used to optimize your site’s copy.
Smart businesses aren’t just waiting for visitors to find them; they’re using social media to drive people to their sites. But how effective is this tactic for you? Google Analytics can tell you, so you can judge whether you’re getting a good return on your investment in social media. It can also tell you exactly which links in your marketing emails have brought in traffic and conversions, so you know what’s working for you and what isn’t. With this sort of detail, it can then enable you to calculate how much you’re spending on marketing per customer per channel, helping you target your spending as effectively as possible and reduce your customer acquisition costs.
Once visitors have found your site, how are they using it? Which pages are clicked on most? Which are leading to successful sales or conversions? Flow Visualization reports enable you to see and analyze the path a visitor takes on your site, so you can see where they came from, which pages they moved through, and where they left the site. You can use In-Page Analytics to see in detail what people are doing on a page, and use Event Tracking to measure activities like downloads, video and animation plays, gadgets and Ajax embedded elements. All this enables you to build up a detailed picture of what your users are looking for and what they like – and in turn what is making you money.
On the technical level, you can use reporting tools like Site Speed, Alerts and In-Page Analytics to improve the working of your site and identify potential glitches such as slow-loading pages, poorly-placed content and excessive load that could lead to poor site performance and a frustrating experience for visitors.
You can also build up a picture of who your visitors are. Audience reports provide insight into their age, gender, geographical location and language, as well as their interests (notably, the categories of websites, apps and videos they’ve visited). You can see whether they’re new or returning users, how often they’ve visited before and how recently.
You can analyze the custom data you’re most interested in. This enables you to find out, for example, whether the users who view videos on your site are the ones who go on to purchase more of your products. All this will help you understand your users’ needs better, improve their experience of your site and services, and make sure you’re reaching the audience you want as effectively as you can.
These days, of course, visitors are increasingly likely to be using multiple devices, including smartphones, tablets and games consoles as well as computers. The Universal Analytics features enable you to keep track of visitors even if they’re using multiple devices. This provides you with a more accurate user count and a better understanding of how people interact with your business. In addition, if you’ve produced mobile apps, you can collect data from these and integrate it with your Google Analytics account to get the complete picture.
The depth of Google Analytics means you’re not just looking at static reports, either. Thanks to its interactivity, you can easily set up Content Experiments to test variations of your pages and learn which designs bring you the most conversions. From ad keywords and the photo on your landing page to complete marketing plans, Google Analytics is designed to help you compare different approaches and assess what works best. You can create custom dashboards and reports tailored to the needs of specific teams within your organization, with automated alerts called Intelligence Events to let you know when something out of the ordinary happens, such as a spike in traffic from a particular referring site. You and your colleagues can even add notes – shared or private ones – on the reporting graphs. We’ll show you how to cover all of these scenarios in this playbook.
How to use Google Analytics to measure business objectives
Understanding exactly what works and what doesn’t allows you to target your spending and effort where it’s doing the most good. One commentator, Martin Wong, notes that Google Analytics offers advanced e-commerce metrics such as revenue per paid click, which enables you to see which paid-per-click keywords are making or losing you money. His tip: every month, cut out non-performing keywords. This tip alone, he concludes, can save you 20-30% of your marketing budget.
Another feature, tracking the shopping cart abandonment rate of your e-commerce site, enables you to see where potential customers are leaving your site before finishing a transaction. By identifying and fixing weak spots, you can boost your conversion rate and increase your bottom line without buying more ads.
It’s a simple equation: reduce guesswork and you make your spend work harder for you, substantially reducing your customer acquisition costs as a result.
The odd sale here and there doesn’t on its own make for a sustainable business, but repeat sales to loyal customers will. Google Remarketing enables you to target customers who have already visited your site or bought your products and therefore already know your brand and (with luck) even like it. If they visited your product page but didn’t add items to the shopping cart, you can try a targeted ad offering discounts for those products. If they did buy, you could possibly offer them related products.
You need to ensure that your ads are relevant to what you know the visitors were looking for, a Google Analytics course can give you vital guidance on where to direct your follow-up activity most productively. Because it gives you a great amount of detail about your visitors’ interests and how they behave, you can target your remarketing activity at those most likely to buy (or buy again).
Once you’ve set up Event Tracking in Google Analytics, you can see those visitors who are interested enough to click on a call-to-action button, a drop-down menu or your website’s chat box. Remarketing to these highly-engaged visitors has a much better chance of successful conversion than simply “all visitors”.
The same applies to people who have visited top sold items, spent more than one minute on the page but haven’t purchased yet: targeting these means you are likely to catch people who are very interested in a product, but were still shopping around. In the same way, people often browse on mobile devices when they get the chance, but stop short of buying because they lack the time or because they don’t feel comfortable doing so on a mobile device. Once you’ve identified visitors who used a mobile device to spend some time on your site without buying – particularly if they spend time on a specific product page – you can then remarket to them on desktop devices.
A recent buzzword is “growth engineering” – or “growth hacking” for those who want to sound more maverick. All marketing is aimed at increasing the number of users you have, but traditional strategies – such as social media, email and content marketing – focus on attracting new users to a site or app, while growth engineering focuses on keeping them there and turning the casual user into a brand advocate.
This is important because a large part of growth engineering involves data mining and trend analysis in order to build up a rounded picture of what users are doing at every step of the way. This understanding is then used to evolve strategies for not only retaining these users but converting them into enthusiastic buyers, and testing these ideas in practice.
The key difference between traditional marketing efforts and growth engineering, as one specialist puts it, is the focus on getting users to stay, convert and return time after time to use your product. Growth engineers focus on testable aspects of a site where the data will tell them what is or is not working, so they can optimize the user experience and increase conversions.
You’ll have spotted how Google Analytics is an invaluable tool in this kind of development, thanks to its capabilities for mapping users’ paths through a site in detail, identifying what appeals to them, measuring the effectiveness of “sale funnels” and testing different conversion strategies in a clear and precise way.
1.3 How to get started with Google Analytics
If you’re new to Google Analytics, it can all seem very daunting. Here’s a step-by-step guide to getting up and running in no time.
How to open and set up your Google Analytics account
It’s easy to get started with Google Analytics. Visit the website and – if you already have a Google account – click “Sign in” at the top right, then enter your username and password. If not, click “Create an account” and follow the steps to create an account.
If you have a Google AdWords account and link your Analytics account to it (see the accompanying ‘Google Analytics Playbook: SEO, SEM, Website and CRO’), then you can view Analytics information and reports by clicking the Tools and Analysis tab in your AdWords account and selecting Google Analytics. If you want to access the account options menu, however, then you need to sign in via the Google Analytics website itself.
Google Analytics data sharing settings
When you set up your Analytics account, you need to choose how your data is shared with Google. There are four settings to enable or disable – Google naturally recommends you enable them all, but here’s what each one does:
- With other Google products: shares your Analytics data with other Google services you use, such as AdWords. This makes it possible for example to import your Google Analytics Goals into your AdWords account. Your data may also be shared in anonymous form with other Google services you do not use.
- Anonymously with Google and others: if you enable this, Google says it will use your data for “benchmarking”, which means sharing information in aggregated form and excluding details that could identify your site or users. This is like taking your answers to a survey and using them to build up standard statistics. This makes it possible for Google to provide features such as Industry Benchmarking, which enables Analytics users to compare their own sites with others in comparable industries.
- Technical support: enabling this gives permission for Google’s tech support staff to access your data to help find a solution if you report a technical issue with your account. This is subject to Google’s internal procedures for controlling access to customer-level data.
- Account specialists: enabling this means that other Google staff, not just tech support staff, will be able to access your data and account details.
The last option is an interesting one. Google says it wants access in order that “they can find ways to improve your configuration and analysis, and share optimization tips with you.” What this means is that if you upgrade to an Analytics 360 account, you can ask your “sales specialist” for specific tips on configuring and using your account to best advantage; if you have a Standard account, you can expect “improved marketing communications that offer usage suggestions” based on what Google has observed you are doing. Google’s support documentation makes it clear that Google staff could look at your actual site metrics information, not just your overall Analytics setup and preferences:
“Your Google Analytics sales team could help you find ways to improve your advertising spend, for example, by offering recommendations based on an analysis of your keyword performance. It is always worth considering the option to enrol on a Google Analytics seminar to ensure you’re on top of your analytics game. The Google Analytics marketing team could suggest ways to improve acquisition or other strategic improvements through a monthly email performance report.”
How to tweak Google Analytics data sharing settings
1. Access admin settings
Sign into your Analytics account – you need to be the account administrator – and then switch to the Admin tab.
2. Select site
If you’ve set up multiple accounts, go to the Account column, and use the drop-down menu to select the account you want to edit. Click Account Settings.
3. Enable changes
If you make changes to any settings – see immediately below for what’s available – click Apply to save your changes.
How do you set up a Google Analytics property?
In Google Analytics, a “Property” is a website or app that you tell Analytics to track as a distinct entity – in practical terms this means it has its own unique Tracking ID. You can set up to 25 multiple Properties within one Analytics account so as to track them separately. You can also associate multiple sites or apps with a single Property ID – note that if you do, you can then use the Views and filter options in Analytics to organize and access information about them separately.
Ultimately, how you set things up depends on how your business is organized and what your long-term reporting goals are. If you have multiple sites that are part of the same business unit and share the same strategic goals (and hence the same broad measurement requirements), it can make sense to include them within one Property in Analytics. Think this through as part of your initial strategy and implementation plan.
If you wish, you can add Properties later, change a range of settings for existing Properties, or delete Properties. Note that if you delete a Property, all the data and reporting Views associated with it will also be deleted and can’t subsequently be retrieved.
You set up an initial Property when you set up your Google Analytics account. To add one later, or edit or delete an existing property, you need to proceed as follows:
How to add, edit or delete Google Analytics properties
1. Access Admin tab
Assuming you’re the account administrator, sign into your Analytics account, then click Admin in the menu bar at the top of the page.
2. Select account
If you’ve set up multiple accounts, go to the Account column, and use the drop-down menu to select the relevant account.
3. Add or select Property
In the Property column, use the drop-down menu to select the Property you want, or add one by selecting “Create new Property” and continue as below.
4. Edit or delete Properties
Edit a Property by clicking Property Settings, then tweaking the settings you want. To delete a Property, click Property Settings followed by “Delete this Property” and confirm when prompted.
When you’re setting up a Property at any time, you must first specify whether it’s a website or a mobile app before giving it a suitably identifiable name – this is used exclusively within Analytics, so make it as descriptive as you want. For a website, you then enter the homepage URL, making sure it’s formatted correctly as follows:
Make sure you include the http:// or https:// and don’t add a trailing slash or any other characters afterwards.
Next, select an Industry Category. Finally set the Reporting Time Zone, which simply determines when Analytics treats each day as beginning and ending: if you choose GMT, for example, then a visit to your site at five minutes past two GMT on Monday morning is recorded as a visit on Monday even if the visitor is located in New York, where it’s 9:05pm on Sunday night.
Any new Properties you create will be Universal Analytics Properties, even if your account is an old-style Google Analytics account (more about this shortly).
Where to get Google Analytics tracking codes
The manual approach is straightforward enough for anyone familiar with HTML, but the code does need to be added to every page of the site. If you manage your site using some form of CMS or generate pages dynamically from templates using a technology like PHP, then it’s relatively simple to add the script to your master page headers.
How to set up and use Google Tag Manager
1. Go to Google Tag Manager to create an account (or to access an existing account).
2. Create a container for your site in the account.
3. Add the container snippet to your site.
4. Migrate any hard-coded tags (such as AdWords or DoubleClick tags) from your site’s source code into Google Tag Manager.
In the case of apps, you need to download the Google Analytics SDK and use its Developers Guide for iOS or Android to learn how to integrate the SDK in your app. Google Tag Manager can be used to manage and update your apps, and its container-based paradigm is particularly useful if you’re producing apps that might be configured differently according to screen size, device or language. In the case of other digital devices – such as information kiosks, games consoles or appliances – you use the Measurement Protocol to collect data. Google advises that “Only experienced developers should set up the Measurement Protocol.”
After you complete the basic setup, you can customize your tracking code to collect data that isn’t tracked automatically, such as transactions and product purchases (e-commerce) or user behavior across primary domains and sub-domains (cross-domain tracking). Google also recommends setting up Event Tracking, which enables you to track interactions like videos and button clicks.
What is Universal Analytics?
Universal Analytics is Google’s new standard protocol for data collection and reporting. It was introduced on a trial basis in October 2012 and declared “out of beta” in April 2014.
Universal Analytics delivers some significant advantages over its predecessor (usually referred to simply as “classic Analytics”):
- Track a user across multiple devices and sessions using User ID, giving you a more accurate measure of real user numbers and engagement.
- Customize your tracking code more easily. Cross-domain tracking for websites, in particular, is dramatically simpler and more accurate.
- More configuration options, including custom dimensions and custom metrics. Among other things, you can for example change how long a “session” is from the default 30 minutes to whatever duration suits your site, and customize how organic search sources (search engines) are treated.
- Set up Enhanced Ecommerce reports to analyze users’ shopping and purchasing behavior, evaluate the success of internal and external marketing, and measure the economic performance of specific products.
If you create a new Property in Google Analytics, the tracking code that’s generated now automatically uses Universal Analytics. If you have older Properties and tracking codes, you can check to see whether they’re using UA.
If you’ve been using Google Analytics for a while and they’re still using the older codes, simply follow the two-step Universal Analytics Upgrade process to upgrade from classic Analytics: you first transfer the Property to UA (in some cases Google might do this automatically) and then (but only then) upgrade the Property’s tracking code.
Universal Analytics works differently in some respects from the old system, so if you’re upgrading Properties there are some usage guidelines to be aware of. You might need to update your site terms and conditions or privacy statement, and ask users to give fresh consent.
For long-term Analytics users, Google has provided answers to some FAQs about upgrading.
How to build your analytics team
We’ve mentioned that you should develop a Google Analytics strategy, measurement plan and implementation plan before actually plunging into using Analytics. The bigger your business, the more people are likely to be involved in all of these steps. In its support documentation, Google talks about “the skills you need on your analytics team”:
You need someone who understands what the business objectives are and the strategies used to support those objectives. You also need someone who understands what analytics can do. Finally, you need someone with technical skills who can implement an analytics tool.
In a small business, you might have just one person – possibly even yourself – who combines all these skills. Even in the largest organization, it will be enormously helpful if there’s an overlap of skills, so that for example the IT team responsible for deploying Analytics into your sites and apps all have a personal understanding of the business’s objectives and can therefore appreciate at every moment that what they do must align with those objectives and serve them.
In practical terms, the biggest challenge seems to be finding staff with a technical grasp of implementing and using Google Analytics. In a US survey conducted in early 2014, marketing organizations reported that their biggest “digital marketing talent gap” was in analytics: 76% felt that analytics was an important or very important skill for applicants to have, but only 39% said that their current team was stronger in this respect than their competition, which means that the “talent gap” is a huge 37%. This was by some way the biggest digital talent gap in all the areas of expertise mentioned, despite the fact that – more than any other skill – analytics was rated as important or very important by the most respondents.
Should I hire Google Analytics Staff
If you’re not lucky enough to have a Google Analytics expert on staff, you might need to recruit one. In roughly the last three months of 2104, under 1% of all permanent IT jobs advertised across the UK specified Google Analytics skills, although over 7% of jobs in the Business Applications category did, which represents a fairer indication of the demand. For permanent positions citing Google Analytics, the average salary offered was £37,500. Some 10% offered a salary of more than £55,000, but the average excluding London was £32,500.
Analytics consultant Michael Loban advises that interviewing for an analyst is an art. “There are plenty of people who are good at using tools or printing out charts,” he says, “but this is not what you are looking for. You are looking for the person who can do all of that before lunch and then focus on maximizing outcomes.”  He suggests that you test whether the candidate can explain web analytics to the least analytical person in the office: “Analytics and insights have no value if the person who has them lacks the ability to share them.”
Test their familiarity with the specifics of Google Analytics: ask what their favorite reports are, or what features they don’t like. Give them a report, a chart, a graph or an analysis and ask them to explain it: “It is one thing to crunch a report, a vastly different thing to analyze it.”
Should I outsource to a freelancer or consultant?
Instead of hiring, Analytics requirements are often outsourced – particularly by smaller firms. This ideally gives you access to someone up-to-date with the rapidly evolving tools used in the industry and “experienced in separating the signal from the noise.”  It can also be less expensive than taking on a permanent employee when your needs are relatively modest – however, we’ve seen that identifying your measurement needs requires close familiarity with the objectives and technological infrastructures of your business, which only someone within the business is likely to have. With this in mind, you’ll likely need to keep analytics strategy and planning in-house.
How to decide if to outsource Analytics
- Local or offshore: business reorganizations, mergers and acquisitions are common in offshoring operations, so you might sign on one company and find it’s taken over by another associated with one of your competitors. Ask also about the outsourcer’s analyst retention rate – you don’t want a continual turnover of analysts working with your data.
- A matching culture: this helps ensure a productive relationship. If you’re a small local company, will a large outsourcer used to working on a global scale be the best fit for your firm?
- Data security: understand what the outsourcer will do to protect your data, from encryption to internal access policies. Is there any risk of your information leaking to your competitors?
- Exit strategy: think ahead to the possibility that your business will grow (thanks in part, no doubt, to good decision-making informed by good analysis) and you’ll want to bring business analytics functions in-house.
- Relationships: just as if you’re outsourcing IT, you need both specialist skills and a productive relationship. Even if you’re a smaller business, it might be wise to have an in-house expert or independent consultant as the relationship manager.
How to connect Analytics to Google Webmaster Tools/Google Search Console
Google Webmaster Tools/Google Search Console is a completely separate free service that enables you to analyze how Google Search views your site, optimize the site’s visibility to Search and consequently help get the best possible ranking in Google Search results. Webmaster Tools can identify internal and external links and alert you to broken ones. Perhaps most usefully, you can view what keyword searches led to the site being listed in Google’s search results, and the click-through rates from these.
Webmaster Tools is much more modest than Analytics, but does complement it: for example, by listing traffic from each keyword separately. Note, though, that Webmaster Tools covers Google search only, not other search engines such as Bing or Yahoo. Plus, as we’ve noted, Analytics covers apps and other devices as well as websites.
They’re both Google services, so it’s not really surprising that they can work together. If you want to add a site to Webmaster Tools, you need to verify that you own the site. Two possible ways to do this are by using your Google Analytics tracking code or Google Tag Manager container snippet.
That’s not all, though. If you’re using Webmaster Tools in addition to Analytics, you can integrate the two services, which means you can view the data from Webmaster Tools in your Analytics reports and vice-versa. Not only is it convenient to see complementary information in one place, but integrating the two services enables useful new functions, such as using Analytics filtering capability to filter by a keyword string. More importantly, you won’t see any reporting data under the Search Engine Optimization section in Analytics (Acquisition > Search Engine Optimization > Queries) until you enable Webmaster Tools data sharing for your web Property in Analytics.
There are a couple of limitations to bear in mind. You can only connect one Webmaster Tools account to one Analytics Property, and a Webmaster Tools account can handle only one sub-domain. So if your Property includes several sub-domains – or for that matter several domains, multiple sites or even mobile apps – then your Webmaster Tools data is going to relate to only a part of the Property you’re tracking. Conversely, if you have multiple Webmaster Tools accounts to monitor multiple domains or sub-domains, it’s not possible to feed the information from more than one of these into one Analytics Property.
In addition, if you connect Webmaster Tools with Analytics, you cannot view Search Query data by landing page. If this is a metric you often use, then integration might not be worthwhile for you.
Linking Webmaster Tools/ Google Search Console to Google Analytics Web Property
1. Log into Webmaster Tools, go to the home page, and click “Manage site” next to the site you want.
2. Now click “Google Analytics Property”, and select the web Property you want to associate with the site.
3. Click Save and you’re done.
Note that there might be some discrepancies between the data from Webmaster Tools and the data from Analytics. For example, Analytics records visits only to pages that have the tracking code correctly configured and embedded in them. This should be all the pages on your site, but accidents happen.
On the other hand, visits to pages without the Analytics tracking code will still be tracked in Webmaster Tools if users reach them via search results or if Google crawls or otherwise discovers them. Finally, be aware that the two services define “keywords” differently: the Keywords page in Webmaster Tools displays the most significant words Google found on your site, while Analytics uses the term “keywords” for both search engine queries and AdWords paid keywords.
How to use Google Analytics for mobile apps
We’ve mentioned that Analytics can deal with apps as well as web Properties. Google has its own step-by-step guides to setting up Analytics for mobile apps, but be aware that doing so requires some tweaking of your app code.
What data can Google Analytics Report?
- Audience report: user location and languages, devices used, app versions, how often users continue using your app.
- Acquisition: how many are new users, what marketplace they got the app from. Google Play Referral Flow tracks a user’s entire path through Google Play, while the AdWords report traces how users interact with your app if they started by clicking on your AdWords ad.
- Behavior: explore in detail how users interact with your app including the total number of screens seen per session, how often users return to the app and for how long, and much more. Note that many of the Behavior reports require additional setup in the app tracking code that should be completed by your developer.
Take note of Google’s advice on Best Practices when implementing Analytics for mobile apps: if you have different apps – or the same app on different platforms – use separate Properties to track them. However, you should track updates and versions of the same app in the same Property – Analytics can distinguish them, so can report on them separately if required.
How to set up Google Analytics mobile app analytics
The process for setting up Mobile App Analytics is straightforward enough: first set up a new app Property in your Analytics account, then download the Google Analytics SDK for iOS or Android and implement the app tracking ID. If you’re using Google Tag Manager, set up the new app Property, then create a container for your app in Tag Manager. To do so, download and implement the Google Tag Manager for Mobile SDK, and finish by adding the Analytics tag to your container.
2.1 How to develop a Google Analytics strategy
No matter how freewheeling and inventive your business may be, you need some goals or even just a vision of where you want to go. Read on to see how to fit Google Analytics into your overall strategy for achieving those.
What is a Google Analytics strategy framework?
It’s important to work with a clear idea of what you want Analytics to do for you – otherwise you’ll simply end up with a mass of facts and figures. Google Analytics can give you a whole range of measurements, but what exactly do you want to measure? If you’re in the market to buy a new car, it can be fascinating to find out its engine capacity, top speed and fuel consumption, but if the vital initial criterion is whether it will fit in your compact-sized garage, you really need to know its dimensions first.
In other words, you need to decide what questions you want answers to, and this in turn will be determined by your business objectives. What are your end goals or “outcomes” and how do you measure whether you’ve achieved them?
What common business objectives can we measure using Google Analytics?
- Ecommerce sites: an obvious objective is selling products or services.
- Lead generation sites: the goal is to collect user information for sales teams to connect with potential leads.
- Content publishers: the goal is to encourage users to engage and visit frequently.
- Information or support sites: helping users find the information they need quickly will be of primary importance.
- Branding sites: the main objective is to drive awareness, engagement and brand loyalty.
In the online world, these are the five common business objectives, depending on the nature and purpose of the site.
There will be key actions or events on your site or app that match each of your objectives. If a user makes a purchase on an e-commerce site or submits a contact form on a lead generation site, then the objective has been met, and Google refers to these as “macro conversions”.
There might also be steps along the way to an objective. Take an e-commerce site, for example: adding an item to the shopping basket, or signing up to receive an email coupon or a new product notification. The user hasn’t fully reached your main objective but is clearly coming closer. Google calls these “micro conversions”, and emphasizes that it’s important to measure these as well as macro conversions so that you can understand the whole process your users are going through. This is especially important if they don’t make it all the way to the ultimate objective, so you can identify what steps in the process might need examining.
The key to using Google Analytics effectively, therefore, is deciding upfront what your key objectives and desired outcomes are, and how you can measure these. Then, instead of trying to measure everything possible, create a simple framework aligned with your desired outcomes and objectives.
What are the objectives and benefits of goal-setting in Analytics?
Google suggests the following five steps to design your measurement plan:
We’ve already touched on common business objectives, but of course many businesses and their sites combine several kinds of objective – building brand loyalty and providing information about products as well as simply selling them, for example. For this reason, it can be helpful to take things up a level and formulate an umbrella mission statement.
Noting the specifics of what the business does then constitutes the second of these steps. Working from the mission statement downwards also enables you to set to one side any activities that may be incidental to your business goals and therefore don’t need to be included in your measurement plan. (Of course, you may then wish to separately reassess the value of these activities, too!)
The critical third step is then to decide what metrics will indicate how well your business is achieving each of its objectives. If you’re selling products, for example, these will be relatively straightforward measures such as how much revenue you’re generating, or the average value of each transaction, but don’t forget those micro conversion metrics – it’s important to measure not just the bottom line but every step that helps along the way (particularly so that you can identify the weak link in the chain, if there is one). If, for example, your business also has physical as well as online stores, you’ll want to factor in how many times your site’s store locator is used, or how many times users print out a coupon for in-store use.
Choosing appropriate metrics can take some thought and imagination, particularly in the case of other kinds of sites and objectives. To measure user engagement on a content site, for example, you’ll want to consider how frequently users visit, but also whether they share your brand content on social networks. On a support site, fewer clicks rather than more clicks might correlate to happier customers, so you might wish to build a simple feedback form into your pages to measure user satisfaction more directly. For PR and brand-building sites or activities, there’s a helpful guide to Valid Metrics for PR Measurement produced by AMEC, the International Association for Measurement and Evaluation of Communication, the global professional institute for agencies and practitioners who provide media evaluation and communication research.
Once you’ve defined the key performance indicators (KPIs) you want to measure, you need to decide which segments of data are important to measure. For example, you’re likely to be investing in different marketing channels – search, email and social marketing as well as display – and you’ll want to know how much value you’re getting from each of those. You might also look at how much of your business comes from repeat customers and how much from new ones, so you can see where there are opportunities for driving more customer loyalty.
Finally, you need to evaluate your KPIs against the targets you’ve set, so you can assess whether the business is doing well or doing poorly. Google Analytics makes it simple to incorporate targets and even include data from outside sources such as point-of-sale devices for a rounded picture.
What is a business analytics implementation plan?
In a nutshell, though, an implementation plan simply requires working out what features of your analysis tool – Google Analytics, in this case – you’ll use to capture the data you need to measure. For most implementations, you’ll start with the standard Google Analytics tracking snippet, then use Goal Tracking and the e-commerce module, if relevant, to track the KPIs you’ve identified. To track marketing campaigns, you’ll use Campaign Tracking and AdWords Linking. We’ll look in detail at these and other features as we go along.
Finally, it goes without saying that you’ll inevitably review and revise your measurement plan, and consequently your implementation plan, as time goes on. Both your business requirements and your technical environment can change over time. Your site might evolve, meaning you have new activities to keep track of, or you might launch new apps and services. You’ll want to refine your implementation to keep your data current and useful. As Google puts it, “The measurement planning process should be cyclical, if not continuous.”
How to integrate Analytics with your marketing strategy
There’s another sense in which the process is cyclical. Measurement isn’t just about reporting: you should be thinking about how to use measurement to give you insights you can act upon in your planning and decision-making. Use the data and analysis you have to continuously re-evaluate what’s working and what you need to change.
This means, conversely, that you should plan your marketing activity around tangible, measurable goals. There’s not much point creating a Facebook page just because your competitors are doing it, but if you do so with a specific goal – whether it’s growing a follower base, generating leads or actually boosting sales – then you can more effectively measure your progress.
Google Analytics can help in all these cases, whether it’s a simple measure of how many site visitors are referred by a given social network, or an assessment of how engaged these visitors are (using indicators such as how many pages they visit per session) and on to conversions – whether they sign up to give you leads or actually make a purchase.
And remember that Google Analytics is brilliant for testing as well, so don’t hesitate to try out new and alternative strategies as you identify the need for them – you’ll be able to assess pretty quickly how effective they are. As Google explains it, “Data can be the driver of a continual improvement process for your business.”
It can be summed up thus: raw measurement feeds into reporting, giving decision-makers the information they need in digestible form and subjecting it to analysis, which identifies trends, compares your company’s performance with a benchmark or a competitor, and makes it possible to identify problems or shortcomings. The next phase is to implement and test possible solutions. Finally, repeat until golden brown – or rather, keep applying the lessons you’re learning to keep improving your results and growing your business.
3.1 Google Analytics home dashboard: An overview
There’s so much on offer in Google Analytics that it can feel quite daunting when you first encounter it. Worry not, because here’s our comprehensive guide to getting around Analytics and getting the best from it.
Analytics Home dashboard: An overview
When you log into your Analytics account, you should see your account Home page; switch to the Home tab if not. By default, this screen shows you a list of all of your Properties, with a handy overview of their performance – click “Show Sessions” if session information is not shown.
If you’ve set up a lot of Properties, you can filter the view by only selecting those you wish to view: click the star to the left of a site to select it before clicking the star button next to “Show” at the top of the list hide all unselected items from view. Remove items from view by clicking their star button again, or click “All” next to Show to view all Properties again.
Clicking the name of a Property allows you to navigate to its own screen, but before you do that, let’s take a closer look at what this overview page can tell you.
By default, the overview page displays information for the last 30 days. Click the date range to select a custom period or to compare the currently displayed period with a previous one. The percentage change in each metric will be shown for easy comparison.
A session is a fundamental concept of Analytics, and is used as the basis for various metrics and reports. It’s like a single visit, which might contain lots of interactions – clicks, page views and events – or just a single page view. By default, a session times out after 30 minutes of inactivity – you can set a custom time if you prefer – or at the end of the day, as defined by your site’s time zone setting.
Average session duration
This is a simple measure of how long users are spending on the site. Remember, though, that it won’t tell you how active or engaged they are while they’re using the site.
This metric measures the percentage of sessions that begin and end on the same page without the user interacting with the page. This isn’t necessarily a bad thing – you have a single-page site, for example, of it might indicate that your visitors can quickly find the information they’re looking for and don’t need to search elsewhere on your site. However, for most business sites it’s not desirable because it demonstrates that users aren’t selecting products and taking them to the next stage of the purchase process. Note that if it’s a single-page site, it won’t bounce rate won’t indicate if users are viewing video content or otherwise interacting with the page: to find out, you need to implement other kinds of tracking, such as Events. If you are experiencing a high bounce rate, it maybe worth considering how to reduce this with an online Google Analytics training from Imparture.
Goal Conversion rate
You can set different types of Goals for a site, and these can relate to any of four things:
- Duration: a session lasts a specified time or longer.
- Pages/Screens per session: a user views a specified number of pages or screens.
- Destination: a user arrives at a specified location in the site.
- Event: a specific action occurs, such as a button click or the play of a video.
The first two types are straightforward to define, while Destination and Event Goals require some more details to set up. Once you’ve set up any type of Goal, however, the overview screen will display what percentage of sessions included the successful completion of a Goal (a “conversion”).
3.2 Reporting tab: An overview
The Reporting tab gives you access to your Google Analytics reports and summary displays called Dashboards. It’s probably the most important tab, because it’s where you’ll look for detailed metrics and reports to help make sense of them. You can view different reports using the navigation panel down the left-hand side. Here’s a comprehensive look at how it works.
What is the Google Analytics Dashboard?
A Dashboard is a collection of “widgets”, which can be thought of in a similar way to the Live Tiles found on a Windows Phone or Windows 8 Start screen. They’re basically mini-reports, which can display data numerically or in the form of tables, pie charts, maps and timelines. Some can even give you a constantly updated real-time display of metrics like the number of users on your site, a map of their locations, and so on.
You can select and customize the widgets you want in a Dashboard to see the information you decide is most important – in other words, create customized views of your data. This enables you to pull out and view specific subsets of your data without having to navigate through a lot of standard reports – but if you do want to see more detail, you can click a widget title to open the complete underlying report.
The default Dashboard, named “My Dashboard”, includes a timeline showing number of users, a geo-map of sessions, a table of sessions by browser, and timelines for bounce rate and goal conversions. You can customize it by rearranging the widgets, adding new ones, removing unwanted ones or filtering the data displayed. You can also create additional Dashboards by clicking “+New Dashboard”.
How to manage the Analytics Dashboard
- To change the name of a Dashboard, simply click its title.
- To adjust the date range or compare two date ranges, click the date range displayed at the right-hand side to open the date picker.
- To add widgets, share, export, customize or remove the Dashboard, click the appropriate pop-up in the action bar beneath the name of the Dashboard.
- Add or remove segments by clicking the pop-up beneath “All Sessions” (the default).
- To open a linked report, click the widget’s title. Note, this works only if the widget is linked to a report.
- To rearrange widgets on the page, drag them by their title bars to wherever you want. If you want to edit or delete a widget, mouse over its title bar without clicking and a set of controls will appear – click the edit icon (the pencil) to configure it, or click the X to delete it. This latter action can’t be undone, so use with caution.
- To refresh the Dashboard’s data, click “Refresh Dashboard” at the bottom-right. This updates the display of “standard” widgets; real-time widgets constantly update themselves automatically.
How to add widgets to the Google Analytics dashboard
There are several ways to add widgets to a Dashboard. The first – and most obvious – method is to simply click the “+Add Widget” button, select a type of widget, configure it and click Save.
Alternatively, click “Add to Dashboard” beneath a report title when you’re viewing your standard reports – select the Dashboard to add it to. This can be an existing one or a new one – to base a new widget on an existing one click “Clone widget” (bottom right of the Add a Widget window), then edit it.
You can also download ready-to-go Dashboards from the Google Analytics Solutions Gallery: click Dashboards in the panel at the left-hand side to see what’s available. Or simply take one of a Google Analytics online course to further discover about widgets.
How to add shortcuts to the GA dashboard
Shortcuts work in much the same way as those you find on your PC or Mac, giving you direct access to customized reports you’ve previously saved shortcuts to. Their main benefit is that they open the exact configuration that you saved of a report, so you can return to a customized view displaying specific Segments and so on with a single click. Note that one setting isn’t saved with a shortcut: the date range.
To create a shortcut, open the report you want, configure it as desired, then click Shortcut in the menu bar above the report. To edit an existing shortcut, click Shortcuts to view your list of shortcuts, click the report you want to open, make any changes you want, then click Save in the menu bar to save the new configuration. Finally, to remove a shortcut, click Shortcuts, then click Overview, open the Actions menu for the relevant shortcut, click Delete and then confirm when prompted.
What are Google Analytics intelligence events?
Google Analytics Intelligence constantly monitors your site and generates an alert whenever it detects statistically significant variations in usage or traffic metrics. If you’ve linked Google Analytics to AdWords and have auto-tagging enabled, it will also generate alerts whenever there’s a significant change in traffic from AdWords.
You can also set up custom alerts for more specific events – traffic from Paris-based visitors dropping by more than 20% over a day, say – and even opt to receive an email alert automatically when a specified event occurs. In any of these cases, you can click Intelligence Events to view the relevant reports.
There’s an Overview summary of all automatic and custom alerts that occurred within the active date range: click the Automatic Alerts or Custom Alerts tab to filter for one or the other; click Details to see a graph of the data over time; click “Go to Report” to view the full report for a specific dimension.
In addition, you can view Daily, Weekly and Monthly events reports. Tick the box below the bar graph for each type of alert you want to see: Custom, Web Analytics (automatic) or AdWords (automatic). If there are many alerts, use the Alert Importance slider to adjust how many are displayed: drag it towards High and the report will display only big variations; move it towards Low to view alerts representing smaller variations. Click a bar in the graph to see the alerts for that day, week or month. Click the graph button for a specific dimension in an alert to see a graph of that data over time; as before, click “Go to Report” to see the full report for that dimension.
To create a custom alert, click Reporting > Intelligence Events and navigate to any of the reports we’ve just mentioned (Overview, Daily, Weekly or Monthly). Below the bar graph, under Custom Alerts, click “Create a Custom Alert”. Enter a name for the alert, select which reporting Views you want the alert to be visible in, then set the period you want to monitor (a day, a week or a month). If you want to receive an email when the alert is triggered, tick the box; you can optionally specify other people to receive emails as well. Next, specify the Alert Conditions by selecting a dimension and setting the “alert me when” parameters – for example, “alert me when Landing Page bounce rate increases by more than 20% compared to previous day”. When you’re happy, click “Create Alert” and you’re done.
To edit a custom alert at any later point, click “Manage Custom Alerts” instead of “Create”, click the alert name in the list of custom alerts, make the changes you want, and click Save Alert. To delete a custom alert, find it in the list of custom alerts and click Delete instead.
What are real-time events in Google Analytics?
Click Real-Time for an up-to-the-minute, constantly updated overview of what’s happening on your site or app. This can be invaluable for seeing the immediate impact of changes to site content, assessing the effect of some social media marketing or a specific promotion, or just getting a snapshot of how the site or app is performing.
What can Google Analytics report in real-time?
There are six reports, all of which display the current number of active users, number of hits in each of the last 30 minutes, and number of hits in each of the last 60 seconds:
- Overview: shows the referral sources for active users, the pages through which these users entered your site and their geographical locations.
- Locations: shows the geographic locations of your current active users, plus how many pages/screens were viewed from each city in the past 30 minutes.
- Traffic Sources: shows the sites or campaigns that referred your currently active users. (This isn’t displayed for apps.)
- Content: (or Screens in the case of apps) displays which pages or screens have been viewed in the last 30 minutes. Click “Pageviews/Screen Views (Last 30 min)” above the table to see the total number of views for each page or screen over that period.
- Events: displays the top 20 categories of events, such as clicks, video plays, downloads or other forms of interaction, if they don’t result in a new pageview, that have taken place on your site over the last 30 minutes. These are sorted by number of users for each one, with the percentage of total users also shown for each. Click an event category to see just the activity in that category; click “Events (Last 30 min)” above the table to see the total number of events in each category over that period.
- Conversions: this section displays a table of the top 20 Destination and Event Goals achieved by current active users during their current session, sorted by number of users for each, with the percentage of total users also shown for each. Click a Goal in the table to see just the activity for that Goal; click “Goal Hits (Last 30 min)” above the table to see the number of conversions per Goal over that period.
3.3 Reporting tab: Audience reports
What are audience reports in Google Analytics?
Audience reports can be one of the most revealing areas in Google Analytics. They provide you with insight into who makes up your audience (the Demographics, Interests and Geo reports), how they access your content (Technology, Mobile), and how engaged and loyal they are (Behavior). The better you understand all this, the better you can tailor your site or app for the needs and interests of your audience.
In all the Audience reports, the display is based on the reporting period you define in the date range drop-down menu. Except for Overview and Users Flow, each report includes a summary sessions graph plus a table chart showing the Acquisitions, Behavior and – assuming you’ve set up Goals – Conversions data for each group.
How to enable demographic data in Analytics
The easiest way to enable Display Advertiser Features for a web Property is to first click Admin, then Property Settings in the Property column. Under Display Advertiser Features, set Enable Display Advertiser Features to ON (and while you’re here, also toggle “Enable Demographics and Interests Reports” to ON); finally click Save. Alternatively, you can update your tracking code directly; you must do this in the case of app Properties. If you’re using Google Tag Manager, you simply tick the box marked “Enable Display Advertiser Features”.
If you haven’t enabled Demographics and Interest reports on the Admin page as above, you can alternatively do so from the Reporting tab: click Audience > Demographics > Overview, and simply click the Enable button above the introductory text. If you haven’t already enabled Display Advertising Features, you’ll now be prompted to do so.
An overview of audience reports in GA
This report features a line graph of the number of sessions over the selected reporting period. Then there’s a display of numbers of sessions, users, pageviews, pages per session, average session duration, bounce rate, and percentage of sessions that are new sessions. (If these aren’t displayed, select them from the drop-down that appears when you click on Sessions.)
Beneath is a data table, with a sub-menu to its left with which you can select what the table displays. By default, it shows the top ten languages, the number of sessions in each language, and the percentage of sessions this represents. The options in the sub-menu are to show the top ten Demographics (Language, Country and City), System info (Browser, Operating System and Service Provider), and Mobile (OS, Service Provider and Screen Resolution – more accurately screen size). Alternatively, you can click the appropriate item under Audience in the main sidebar on the far left to access the full reports.
What are audience demographic reports in Google Analytics?
Demographics offer a choice of three views: Overview, plus specific reports for Age and Gender.
The Overview display breaks down your audience by age groups and gender. You can access the full age and gender reports by clicking the link in each chart or from the sidebar. When you drill into an age group, you see the breakdown by gender; drill into one of the age groups within this and you then see a breakdown by interest. Bear in mind that Google does not report users aged under 18, and that age and interest information is extrapolated from visitors’ previous online behavior and should not be taken as infallible. (Find out more in the ‘Google Analytics Audience Segmentation’ playbook)
What are interest reports in GA?
Find out what your audience’s interests are. As with demographics, the Overview page shows a summary of the other sections, which are: Affinity Categories, In-Market Segments, and Other Categories.
- Affinity Categories: identifies users in terms of broad lifestyles – for example Technophiles, Sports Fans or Cooking Enthusiasts. These categories are defined to be similar to US TV audience categories.
- In-Market Segments: identifies users in terms of their product-purchase interests.
- Other Categories: provides the most detail. For example, while Affinity Categories includes the broad category “Foodies”, Other Categories includes the much more specific category “Recipes/Cuisines/East Asian”.
Bear in mind that a single session might be counted in more than one category – for example, “Software/Internet Software” and also “Software/Internet Software/Internet Clients and Browsers” – but each session is counted only once in the total at the top of the column.
What are Geographical reports in Google Analytics?
This report shows your users’ Language or Location (country, region or city).
In this way, you can quickly spot areas with higher or lower than average numbers of sessions. Choose a Goal Set or Ecommerce under the Explorer tab and you’ll be able to spot locations with high or low numbers of conversions. If you want to attract more users from an underperforming area, you might consider adding content specifically relevant to that area, or more content in the appropriate language. To convert visitors from such areas to buyers, you could consider marketing campaigns targeted at them, and so on. Both reports include the standard charts showing Acquisition, Behavior and Conversions data, and Location also displays a map of your visitors’ locations, derived from mapping their IP addresses to geographic locations. You can click this to drill down to the country you want to know more about. To begin, select Site Usage on the Map Overlay tab, and in the map, select the metric Sessions from the drop-down menu. Set Country/Territory as the primary dimension (top left of the table), then click an area on the map to zoom in. You can now change the graphical display to – for example – Metro area or City using the primary dimension options again.
How to understand user behavior within Analytics
Find out how many of your visitors are new and how many returning, how frequently they visit, and how long it has been since their last visit. The Engagement report tells you how long they spend on your site and how many pages they visit.
Select, say, Frequency as the primary dimension above the table, then select Goal Set or Ecommerce (under the Explorer tab, upper left of the report). Finally, select the Data view (top right of the table) if not already selected. If you find that the highest number or value of conversions correlates with repeated visits, then you might benefit from focusing on attracting new users back, perhaps with inducements to subscribe to a blog or email marketing list.
How to understand technology usage through GA
Available dimensions here include Browser, Operating System, Screen Resolution, Screen Colors, Flash Version, Network, and Java Support. Select each of these as a primary dimension above the table, then select Goal Set or Ecommerce under the Explorer tab at top left of the report. If the conversion rates are lower than average for any given technology, this might alert you to a site design issue that needs addressing, perhaps your site is relying on a technology that your visitors are not using?
How to understand mobile device usage via Analytics
The Mobile Overview shows you the number of desktop, mobile and tablet users who visit your site. The Mobile Devices category report shows you exactly what devices those visitors use, and their Screen Sizes. Analyzing this data can help you decide whether you should optimize the site for a smaller or larger screen size, or even create a dedicated app for the appropriate platform.
Want to identify long-term mobile traffic trends? On the Overview report, select a long date range and Week or Month view from the top-right of the graph. In the table, tick the box next to each device category, then click the Plot Rows button at the top left of the table. The line chart will now show mobile, non-mobile, and all sessions over time. Hover your mouse over a data point in the line chart to compare traffic levels.
Want to collate traffic levels with other metrics? Select an appropriate Goal Set from the Explorer tab, and look at the relevant metrics in the table. If desired, plot the metric you’re investigating (such as Goal Conversion Rate) by selecting it from the “Select a metric” drop-down menu in the graph.
What is custom variable reporting in GA?
This option is a relic from the earlier version of Google Analytics. The Custom Variables report showed activity by custom Segments that you created yourself by modifying the Analytics tracking code. However, custom variables are not supported in Universal Analytics. If you’re using Universal Analytics (as you now should be), you need to use custom dimensions instead.
The idea is that you can implement custom dimensions to collect and analyses data that Analytics doesn’t automatically track. For example, if you have a form on your website where visitors specify which industry they work in, you can capture and report on this. Custom dimensions and custom metrics require additional setup in your Analytics account and in your tracking code. Once you complete both steps of the setup process, custom dimensions can appear as primary dimensions in Custom Reports, and you can also use them as segments and secondary dimensions in Standard Reports. Note, though, that they do not appear directly here in the Audience reports section.
How to set-up benchmarking reports in GA
Benchmarking enables you to compare your figures with aggregated industry data from other businesses who share their data. This helps you put your performance in context and assess what’s happening across your industry. To see Benchmarking data, you must agree to share your own data. It’s then included in benchmarks, but the data you share (including information about the account from which it is shared) remains anonymous.
If you didn’t enable data sharing during setup (see section 4.1 of the ‘Getting Started with Google Analytics’ playbook), click the Admin tab and, under Account, click Account Settings, tick “Anonymously with Google and others”, and click Save.
You’ll now be able to see benchmark reports under Benchmarking. There are three reports:
- Channels: compares your channel data to the benchmarks for each channel in the Default Channel Grouping: Social, Direct, Referral, Organic Search, Paid Search, Display, Email.
- Location: compares your Country/Territory data to the benchmarks for each of the Countries/Territories from which you receive traffic.
- Devices: compares your Devices data to the benchmarks for desktop, mobile, and tablet traffic.
Use the selection menus (top of each report) to refine the benchmark against which you want to compare your data:
- Industry Vertical: select one of over 1,600 industry categories.
- Size by daily visits: select from seven traffic size classifications to compare your Property against Properties with similar traffic levels in your industry.
- Geographic location: limit benchmarking data to a specific country or territory.
- The number of Properties contributing to the aggregate benchmark data is shown at the top of the report: metrics available include number of sessions, percentage of new sessions, pages per session, average session duration, and bounce rate.
The table displays the percentage by which your Property outperforms or underperforms the benchmark for each metric.
How to understand user flow through Google Analytics
This report is a little different from the other Audience reports: it enables you to visualize the path that your visitors take through your site. Using the drop-down menu at the top left, you can see the flow of users based on language, location, browser, mobile device and similar dimensions.
Follow users from the starting page where they enter through as many interactions as they make or pages they view on your site until they exit. This enables you to compare volumes of traffic from different sources, examine patterns of traffic through your site and identify any problem spots where you may be losing your visitors.
The graphic shows “nodes” and the “connections” over which users moved between them. The first node represents one value of the dimension by which you choose to filter your view, such as Country/Territory (select this in the first column), and other nodes represent a single page or collection of pages (for example, all pages in the “Footwear” directory). A connection represents the path from one node to another, along with the volume of traffic along that path.
The graphic enables you to see the relative volume of traffic to your site by the dimension you choose (for example traffic sources, campaign or browser) and the relative volume of pageviews per page or collection of pages. Mouse over a connection, node or node exit and you’ll see specific metrics for each. Click one to highlight just the traffic through there or to view only that segment.
You can click the arrows or drag with your mouse to pan left or right, and zoom in or out with the Zoom slider. If things are still too jumbled, you can drag the Connections slider to display fewer (or more) connections. By default, data for the last 30 days is displayed; click the date range (top right) to alter this.Read previous article Read next article
3.4 Reporting tab: Acquisition reports
What are acquisition reports in Google Analytics?
The Audience reports tell you about your users, leaving the Acquisition reports to say where they’re coming from: search engines, social networks, website referrals and so on. This is invaluable for assessing which marketing channels are bringing the most visitors to your site. If you have Goals set up, the Acquisitions Overview report will even show you how well each channel drives conversions.
Analytics makes a distinction between source and medium. The medium is how traffic reaches your site: paid or unpaid search, referral from another site, an email campaign (when you explicitly identify this as the medium) or the user entering the URL (example: https://www.imparture.com/course/training-workshops/learn-google-analytics/)directly.
The source is the specific point of origin. This could be the search engine used – such as Google, Yahoo, Bing, etc., or it might be the actual referring site, such as youtube.com). The source can also be one of your newsletters (identified for example as “spring_newsletter”) or the name of an AdWords campaign – although slightly confusingly, if users enter your URL or have it bookmarked then the source is given as “direct” while the medium is “none”. You can use Custom Campaigns (see below) to tag links with your own custom values for Campaign, Keyword, Source, and Medium.
Each report enables you to focus on a different primary dimension such as Campaign, and drill into the data – for example to evaluate each source/medium pair you used to deliver the campaign. In all the reports other than Overview, the Explorer tab lets you take different views of the data for the primary and secondary dimensions you’re using: Summary (the same metrics you see in the Overview report), Site Usage (behavior metrics), Goal and Ecommerce.
What is an overview report in Google Analytics?
By default, the Overview report shows you a summary of sessions by channel, the total number of sessions across the date range, and the total number of conversions across the date range, plus each channel’s relative performance on the Acquisition, Behavior and Conversion metrics.
Acquisition includes number of sessions, percentage of these that are new sessions, and New Users. Behavior includes bounce rate, pages per session, and average session duration. Conversion shows number of transactions, revenue, Ecommerce Conversion Rate, Goal Conversion Rate, Goal Completions and Goal Value.
In conclusion, therefore, the Overview report gives you a quick indication of which channels account for the most users or new users, which channels acquire users who engage most with the site, and which channels acquire users who lead to the most conversions.
What is a channel report in GA?
Organic Search takes you to the Keywords report, Direct takes you to the top landing pages for direct visitors, Referral takes you to your top-referring websites, and Social takes you to your top-referring social networks. Paid Search will show you the paid keywords, search queries and campaigns from which traffic originated, and Email the email campaigns from which traffic originated.
This report is similar to the Overview, except it gives you a graph to go along with the Acquisition, Behavior and Conversions details – click any of the channels to see related standard reports with more details.
If you wish, you can alter the Default Channel Grouping, which is based on the most common sources of traffic, like Paid Search and Direct, but might not suit your specialist needs. If you want to label your traffic in other ways, you can create a new Channel Grouping (recommended) or even edit the Default Channel Grouping. For example, you might wish to separate your specific Brand Paid Search channels from Generic Paid Search, which might perform quite differently.
What is an all traffic report in Analytics?
This report lists your top traffic sources regardless of channel, ranked according to the number of visitors they sent to your site. So, for example number 1 might be a specific search engine, number 2 a referring website, number 3 a specific directory you advertise with, and so on.
Included in the Google Analytics traffic report:
- Source: as explained at the start of this section – for example “google” (the name of a search engine), “facebook.com” (the name of a referring site), “spring_newsletter” (the name of one of your newsletters), and “direct” (if a user typed in your URL or had bookmarked your site).
- Medium: for example, “organic” (unpaid search), “cpc” (cost per click or paid search), “referral” or “none” (direct traffic has a medium of “none”).
- Keyword: The keywords that users searched for may be captured in the case of search engine referrals, whether organic or paid search, but note that when SSL search is employed (as in Google searches), Keyword will have the value “(not provided)”. See the section on the Keywords report
- Campaign: the name of the referring AdWords campaign or a custom campaign that you have created.
- Content: identifies a specific link or content item in a custom campaign. For example, if you have two call-to-action links within the same email message, you can use different Content values to differentiate them so that you can tell which version is most effective. You can use Custom Campaigns to tag links to use your own custom values for Campaign, Medium, Source, and Keyword.
Along with the search engines and campaigns (Sources) that are sending traffic, the All Traffic report lets you see a breakdown of Organic Search Traffic vs Paid Search Traffic (Traffic Type). Note that the latter includes AdWords traffic. Bear in mind that the raw number of users isn’t the whole story: look at the Site Usage figures and you might find that while organic search delivers many times more users, those users view only half as many pages and have twice the bounce rate. Look at the Ecommerce statistics, and you might find that users who arrive via paid search have a much higher completion rate and a higher average value per transaction, meaning that your investment in paid search is succeeding in bringing in customers who are after your products.
What is a referral in Analytics?
This report leaves out search engines and direct traffic, so it shows just the web domains (including social networks) that have referred traffic to your site. Click on any of the domains listed, and some will show you the specific pages that referred traffic. This is helpful if the referral source is a blog, for example, so you can see the specific posts that are sending visitors to your website.
In the past, self-referrals used to be a problem – that is, your own site showing up as the source of a visit. With Universal Analytics, this is now unlikely to occur.
What is a campaign report in GA?
This shows traffic from your AdWords or Custom campaigns: how many users each brings in, how many pages they view, how much they’re spending and so on.
If you find that a campaign is bringing in plenty of users, but a lot of them are leaving after viewing only one page (bouncing), then you might have a problem with the landing page associated with that campaign. If you’re running the same campaign through multiple sources (such as Yahoo or Bing in addition to Google), you can view the data by Campaign and add the secondary dimension Source so you can compare the results from the different sources side by side. The Campaigns report includes all traffic (from campaigns and from other sources), so you can assess whether each campaign is bringing in users with higher completion rates than users who reach your site by other means.
You might see a number of sessions where the campaign is listed as “(not set)”. The majority of these are sessions that have no campaign tagging. Many of these will be traffic from other (non-campaign) sources, but some may be from AdWords campaigns that have some error in the tagging. You can eliminate these statistics by using the Search option above the table: set the parameters to Exclude Campaign Exactly matching “(not set)”.
To use Custom Campaigns, you need to add parameters to your URLs that can identify each separate link in a specific campaign email, for example. Probably the easiest way to do this is to use Google’s URL builder online tool. Then, when users click one of the custom links, the unique parameters are sent to your Google Analytics account, so you can identify the exact links that are most effective.
What is a keyword report in GA?
The Keywords report breaks down the keywords that visitors used to find your site. In theory, this applies to both organic and paid searches. Unfortunately, when SSL is used – as it is in Google Search – the keyword is not exposed and Analytics records only “(not provided)”. So, in practice this report shows you mainly the paid search terms that triggered your ads.
You might find, however, that this isn’t very helpful if the vast majority of these are simply your own company name or your domain. If you prefer, you can click the Admin tab in your Analytics account and exclude these terms, or any others you like, as search terms. Any traffic that finds your site by searching an excluded term then isn’t included as search traffic in your Analytics reports but as direct traffic.
To see your organic search keywords from Google, you can try Google Webmaster Tools (look under Search Traffic > Search Queries) or third-party tools like HitTail. Of course, neither of these can give you the sort of detail that Analytics offers or correlate with conversions, but you’ll at least have an idea of the keywords people are using to find you. (See the Search Engine Optimization report, below.)
How to measure cost analysis via Google Analytics?
This report shows session, cost and revenue performance data for your paid advertising campaigns. It includes any non-Google marketing channels for which you upload cost data, and AdWords (labelled “google/cpc”) if you’ve linked your AdWords and Google Analytics accounts (see section 2.5 of the ‘Google Analytics for SEO, SEM, Website and CRO’ playbook), plus imported AdWords cost data to the View you’re using. The report compares the cost of each campaign with its associated revenue (from ecommerce and/or goal value) to calculate ROAS (Return on Ad Spend) and RPC (Revenue per Click).
The Cost Analysis report is mainly concerned with your non-Google campaigns. The AdWords report shows you data about the visitors who click through your AdWords campaigns. You need to have linked your Google Analytics and AdWords accounts. If you have auto-tagging enabled, AdWords cost data will already be available in these reports by default.
How to measure referrals from social media?
This report gives you information about referrals from social networks. The Overview gives you a summary of conversions linked to social networks and traffic from specific networks. The Social Value graph shows the number and monetary value of conversions resulting from social referrals and compares these with all conversions. When a session from a social referral results in a conversion immediately, this is labelled “Last Interaction Social Conversions” in the graph; if a referral from a social source does not immediately generate a conversion but the user returns later and converts, then the referral is included in “Assisted Social Conversions”.
You can drill down to see specifics in seven additional sub-reports:
- Network Referrals: this report shows you the top social networks driving visitors to your website, with Engagement metrics (pageviews, pages per session and average session duration) for traffic from each. This report is enhanced with off-site data for Google Analytics Social Data Hub partner networks; these include Blogger, Delicious, Digg, Meetup, Reddit and others. Click on a partner network to see the URLs that were shared on that site. Change the dimension to Social Network and Action to see what actions people are taking on the network (for example, a “+1” or “comment” action).
- Data Hub Activity: this shows you how people are talking about and engaging with your site content on social networks. You can see the most recent URLs people shared, how and where they shared (via a “reshare” on Google+, for example), and what they said. Use the networks drop-down to select different Social Data Hub partner networks. Click Conversations to see posts and comments about your content, or click Events to see other social actions (like +1 clicks).
- Landing Pages: shows engagement metrics for each page on your site that received referred traffic. Click one of your site’s URLs in the table to see which social networks sent the most traffic to that specific page.
- Trackbacks: reveals which sites are linking to your content, and in which context. You can see each endorsing URL’s page title and publication date, plus the number of sessions that it sends to your site. Use the More drop-down in each row to view the originating site or your own page that was shared. Use the Filter Pages field to filter by your page URL.
- Conversions: reveals which social network traffic is leading to the most conversions and the greatest value of conversions on your website. Click “Assisted vs. Last Interaction Analysis” (just below the Explorer tab at the top of the report) to compare assists with immediate conversions. Note that you must define Goals and Goal values to see data in this report.
- Plugins: if you have Google “+1” and Facebook “Like” buttons on your site, the Plugins report will tell you which buttons are being clicked and for which content. Google +1 interactions are tracked automatically, but additional technical setup is required to track Facebook and other social plugins. You can do this either in your site code or using Google Tag Manager. Note that Google Webmaster Tools counts differently and updates less frequently, so the number of interactions it records might be different.
- Visitors Flow: this final report displays the path that visitors take through your website after coming to it from a social network. Mouse over a source on the chart (Google+, for example) and select “View only this segment” to focus on traffic from that source.
How to measure search engine optimization (SEO) via GA
The Search Engine Optimization (SEO) reports provide information about Google Web Search queries that have returned URL results from your site. To make these reports available, you need to add your site and verify it with Google Webmaster Tools, then configure SEO reporting within Google Analytics. The SEO reports use four specific metrics:
- Impressions: the number of times any URL from your site appeared in search results viewed by a user, not including paid AdWords search impressions.
- Clicks: the number of clicks on your website URLs from a Google Search results page, not including clicks on paid AdWords search results.
- Average Position: the average ranking of your website URLs for the query or queries. Keep in mind that the most typical search queries will return only a single URL from your site. If a query returns more than one page, the average position is based on the most prominent URL in the search results when only a low number of URLs from your site are displayed.
- Click-through Rate: clicks divided by Impressions, expressed as a percentage.
- Queries: as we’ve noted, Webmaster Tools can discern some of the keywords that people use to find your site; the Queries report shows the Google search queries that generated the most impressions for your pages, along with the number of impressions and number of clicks, plus click-through rates (but not conversions) for each keyword. You can sort by any of these columns. There are three sub-reports:
- Landing Pages: shows you the pages that receive the most impressions and clicks from search, along with their click-through rate and average position in search.
- Geographical Summary: provides a general view of Impressions, Clicks and Click-through Rate by country. You can also select Google Property as a primary dimension to see a breakdown of Google search activity by Web search, Mobile search, Video search and Image search.
3.5 Reporting tab: Behavior reports
What are behavior reports in Google Analytics?
If the Audience reports tell you who your site’s visitors are and the Acquisition reports reveal how they got there, the Behavior reports show what they’re doing on the site. This should help you assess how well the content, design and technologies of the site meet the needs and expectations of users.
What metrics are reported in behavior overview?
- Pageviews: which might include repeated views of the same page – each view is counted as a pageview.
- Unique Pageviews: the number of individual users who have viewed a specific page at least once during a visit. This time, if a user views the same page more than once during the same visit, only the original view is counted.
- Average Time on Page: or optionally on a defined set of pages or screens.
- Bounce Rate: the percentage of single-page visits or the number of visits in which people left the site from the same page they entered on.
- % Exit: the percentage of users who exit from a given page or set of pages.
You’ll find links to reports for top content page URLs, top content page titles, search terms, event categories and AdSense revenue below the overview graphs
What is a behavior flow report?
This report shows the path users travelled from one page or event to the next until they left the site. Use the view type selector at the top of the report to see user movement between Pages, Content Groupings, Events, or both Pages and Events. Note that you must have set up and be tracking Events before they appear in the Behavior Flow report. You must also have set up Content Groupings before they appear in the report. Content Groupings are created at the View level: click Admin, select the View you want, then select Content Grouping.
In this report, page nodes are green, event nodes are blue, and dimension nodes are white. Click a node to highlight or explore traffic through that node, or to see the individual pages or events that are grouped together in that node.
A connection represents the path from one node to another, and the volume of traffic along that path. Click a connection to highlight just that traffic segment through the flow.
Note that in Events view, exits don’t necessarily indicate that the user left the site, only that a traffic segment didn’t trigger another Event. Exits are not shown in the “Pages and Events” view.
You can use the Behavior Flow report to determine how engaged users are with your content and to identify potential content issues or navigation problems. Are there paths through your site that are more popular than others, and if so, are those the paths that you want users to follow?
What does a site content report include?
The Site Content report contains four sub-reports:
- All Pages: this displays the most frequently viewed pages on your site, offering each page’s pageviews, unique pageviews, average time on page, entrances, bounce rate, and % exit. It also displays the page value (defined as the Transaction Revenue + Total Goal Value divided by Unique Pageviews for the page or set of pages), if you’ve defined all the necessary components.
- Content Drilldown: this report allows you to see the top folders of content on your website and the top content within that folder. While it looks similar to the All Pages report, the distinguishing feature here is the ability to see top content sections instead of just top content pages.
- Landing Pages: the top pages on the site by which visitors enter the site. Metrics for landing pages include Acquisition (sessions, % new sessions and new users), Behavior (bounce rate, pages per session and average session duration), and Conversions based on Goals you’ve defined for the site.
- Exit Pages: shows the last pages people visited before exiting the site. These are the pages you want to look at to see whether you can do anything to keep visitors on the site longer.
What is reported within the site speed report?
This report is based by default on a fixed 1% sampling of your site’s pageviews, though this can be customized if you wish.
The Overview report shows a range of metrics including Average Page Load Time (in seconds), Average Redirection Time before a page is fetched, Average Domain Lookup Time, Average Server Connection Time (the time in seconds spent in establishing TCP connection for a page), Average Server Response Time (including the network time from the user’s location to your server), and Average Page Download Time. You might feel that some of these things are influenced by factors outside your control, but some of these measures can alert you to issues you might aim to improve by such means as optimizing content on your pages, reducing the size of images, or cutting the number of widgets and plugins on your pages.
Under the Site Speed metrics, you’ll see quick reports on average page load times based on the browser the visitor uses, the location (country) of the visitor and the page the visitor lands on.
There are three sub-reports with the site speed report:
- Page Timings: displays how long your most-visited pages take to load compared to the overall average load time for the site. Further information is provided behind various tabs, including site usage metrics such as pageviews and bounce rate, technical network and server metrics, and a map overlay displaying different geographic data.
- Speed Suggestions: gives you step-by-step advice from Google on how to optimize specific pages on your website. The PageSpeed Score indicates the extent to which you can improve the load time of a page – the lower the score, the more room for improvement. Note that the score isn’t itself a measurement of the page’s speed, but only the extent to which the speed can be improved.
- User Timings: this report enables you to measure the execution speed or load time of specific elements on a page, including any discrete hit, event or user interaction that you want to track, such as how quickly images load or the response time after a button click. Note that User Timings requires you to implement custom code on your website.
What is a site search report?
The Site Search Overview report displays the overall metrics for visitors who use the search box on your website. Beneath these, you can view quick reports for the terms searched, categories and the pages where visitors initiated a search.
If you provide a search box on your site, use the Site Search reports to find out how successful your users are when they use it to search your site. Metrics include the number of sessions that used your site’s search function at least once, what percentage this is of total sessions, how long users spent on your site after performing a search, and the number of pages they viewed after performing a search (“Search Depth”).
The Usage report breaks down the number of visits where someone used the search box on your website versus the number of visits where the search box wasn’t used. You can quickly see whether having a search box increases or decreases factors like bounce rate, average time on your website and conversions. Metrics for the pages users land on as a result of their search include Acquisition (sessions, % new sessions and new users), Behavior (bounce rate, pages per session and average session duration) and Conversions based on your website Goals.
The Search Terms report displays the keywords entered into your website’s search box. Along with the terms, you’ll find metrics for the total number of searches, % search exits and additional details about visits related to a search term.
The Pages report displays the same metrics mentioned above for search terms, but in this case the metrics are focused on pages where searches originated.
The % Search Exits figure represents the percentage of searches in which the user simply left the site after searching instead of clicking any of the results pages that you offered. A high figure implies that the search isn’t turning up what they’re looking for, so you might need to tune the site search to provide relevant results.
You must set up Site Search reporting for each reporting View. Click Admin, select the View you’re using, then click View Settings > Site Search Settings, and set Site Search Tracking to ON. You need to know the site search parameter used on your site (often “q=”) and set a few other preferences.
What is included within an events report?
In addition to viewing pages, users might interact with other kinds of elements on your site such as Flash, Ajax applets or just videos. The Events reports enable you to track such interactions. You’ll first need to set up Event Tracking code on your site.
The Events Overview report displays a summary of the visitor interactions you’re tracking. Values are calculated based on the event value you specify in your event tracking code. Under these metrics, you’ll find quick reports showing the number of events based on category, action and label (all of which are specified in the Event Tracking code you set up):
- Top Events: reveals the events with the most visitor interaction.
- Pages: shows you the top pages where visitors interact with the events you’re tracking.
- Events Flow: displays the path that visitors take on your website from when they arrive to when they interact with your event. The default view shows event interactions from visitors in specific countries. You can change the view to show event interaction flow from landing pages and other dimensions offered in the drop-down menu above the first column.
How to set-up an Adsense report in Analytics?
To view these reports, you first need to link your Google AdSense account to Google Analytics. Click Admin, then select the relevant Account if you have more than one. Now click your Property, then AdSense Linking and click “+New AdSense Link”. Select the AdSense Property that you want to link with your Analytics Property, click Enable Link, then Done.
Once the data filters through, the graph in the AdSense Overview report shows by default the daily total AdSense revenue for your site. Use the graph to compare metrics. The table below lists 10 AdSense metrics. Click any metric to see the daily values.
The AdSense Pages report displays the top pages on your website that generate the most AdSense revenue. Additional metrics show the number of ads clicked, click-through rates, revenue per thousand impressions and overall impressions per page. By default, the graph displays the daily total AdSense revenue for your site; the table displays total revenue metrics for the date range distributed by page.
The AdSense Referrers report shows you the referring URLs driving visitors to your website who click on AdSense ads. The graph displays the daily total AdSense revenue for your site; the table displays total revenue metrics for the date range distributed by referring domain. In the table, click a domain name to see revenue metrics per referring page.
How to set-up A/B testing experiments within Google Analytics
These reports make it possible to conduct simple A/B/N testing to see which landing page variations perform best at meeting specific objectives – you can test up to 10 full versions of a single page at a time, each delivered to users from a separate URL.
Once you’ve set up and launched your experiment, a random sample of your visitors see the different pages, including the original home page, and you simply wait to see which page gets the highest percentage of users to achieve the desired objective (buying the specified product, for example). When you see which page drives the most conversions, you can make that one the live page for all users.
To set things up, simply navigate to Behavior > Experiments and click “Create experiment”. The wizard will guide you through setting up your first experiment.
What is required for a Google Analytics experiment?
- Different versions of your web pages to serve to your visitors: Google advises that you change only a few elements, but make the changes bold enough to matter, and use high-volume pages so you get a good number of users.
- Adding the experiment code to the original page in each case: so that sample groups of visitors are redirected to the variation pages.
- Goals: these need to have been previously set up in Analytics (see section 3.3 of the ‘Google Analytics for SEO, SEM, Website and CRO’ playbook).
- Ecommerce tracking enabled: if you want to use purchases or other commercial metrics as the objectives.
Each experiment page is measured according to the percentage of users viewing that page who accomplish the objective you set. Note that in addition to Goals, you can use any other available metric as objective for your experiment. For example, you can test which page leads to a decrease in bounce rate, or to the greatest increase in revenue or session duration.
Experiments should become part of your ongoing development routine. Don’t stop after you’ve tested one set of page variations – follow up and keep experimenting!
What is an In-page Google analytics report?
This report gives you a way to see what links your visitors clicked, page by page. You first need to go to Admin > View Settings to enter the URL for the page on which you want the report to launch. Then you can navigate In-Page Analytics the way you navigate your site: click any link on your homepage, and when the new page loads you’ll see the corresponding data for that page.
The links that visitors clicked are shown in a graphical overlay of bubbles. The numbers inside represent the percentage for the metric you chose in the Viewing menu. When you hover over a bubble, you see information about the metric you selected – for example, if you’ve selected Clicks, you’ll see the number of clicks on that link and what percentage of clicks these represent. You also see the destination URLs for the link, and the number of other links on the page that lead to those same destinations. Note that In-Page Analytics reports traffic data only, not Ecommerce data.
The handy Browser Size feature lets you see the portion of each page that is visible without scrolling to a specified percentage of your visitors.
There’s a neat additional option if you’re using the Google Chrome browser: install the Page-Analytics Chrome extension, and you can see some Page Analytics information right there on a page you’re viewing within the browser, in addition to through the normal Analytics reports. The page must be one you’re tracking with Analytics, and the information you can see within the browser is limited to the following:
- Metrics: Pageviews, Unique Pageviews, Average Time on Page, Bounce Rate, % Exit
- Number of active visitors: in real time
- In-page click analysis: where users click on that page.
By default, the information appears in “scorecards” across the top of the page.Read previous article Read next article
3.6 Reporting tab: Conversions reports
What are conversion reports in Google Analytics?
What goals can be reported in GA?
These reports give you an overview of your site’s Goals. As we’ve mentioned, you can set different types of Goals for a site:
- Duration: a session lasts a specified time or longer.
- Pages/Screens per session: a user views a specified number of pages or screens.
- Destination: a user arrives at a specified location in the site.
- Event: a specific action occurs, such as a button click or the play of a video.
You set these up at the View level: click Admin, then a specific Account if you have more than one, then Property and select a View. Now click Goals, then “Create a Goal”. Simply follow the steps, then click Save Goal to finish. You can use a template or create a custom definition. Note that the Goal categories (Revenue, Acquisition, Inquiry, Engagement) are for guidance only, and are not reflected in reports.
You have the option of assigning a monetary value to a Goal during setup. If you’re tracking a transaction or purchase with the Ecommerce tracking code, though, leave the Goal Value blank: the actual value of the transaction will appear in the Revenue metric (not the Goal Value metric), and will come from the Ecommerce tracking code in your shopping cart.
You’re limited to 20 Goals per reporting View. To track more than 20 Goals, create an additional View for that Property, or edit an existing Goal you don’t need any more. Once created, Goals can’t be deleted, but each Goal can be turned off, in which case no data is recorded for that Goal.
After you’ve created and saved a Goal, you can share it with others. To do this, go to the Goals setting under the View column on your Admin page. Click Share to create a URL you can share with others – note, only the configuration information is shared; your data remains private. You can import Goals that other people have created: visit the Google Analytics Solutions Gallery and click Goal in the menu at the left-hand side.
The Goal Flow report, like the other flow reports we’ve seen, shows the path that your visitors travelled through on their way towards a Goal conversion.
What is an Ecommerce report within Analytics?
These reports make it possible to track which products your customers buy, in what quantity, and the revenue generated by those products. In addition, if you look under Transactions you can find out the revenue, tax, shipping cost and quantity information for each transaction; search under Time to Purchase to find out the number of days and sessions it took a customer to make a purchase, starting from the most recent related campaign through to the completed transaction.
What is an enhanced Ecommerce report?
This report provides details of the customer’s path to purchase, such as when customers added items to their shopping cart, started the checkout process and completed a purchase. Importantly, Enhanced Ecommerce gives you the ability to identify segments of customers who are falling out of the shopping funnel.
The Shopping Analysis reports give you a detailed look at how users engage with your content in terms of viewing products, and adding or removing them from shopping carts, along with initiating, abandoning, and completing transactions.
The Overview and Product Performance reports include data for the revenue and conversion rates of your products, how many products the average transaction includes, the average order value, refunds you had to issue, and what proportion of views of product information pages led to purchases. A basic implementation of Enhanced Ecommerce tagging gives you data for individual products, but you can also add categories and brand groups too, as appropriate.
The Affiliate Code report enables you to track revenue, transactions and average order value associated with defined affiliate sites that drive customers to your site. You can do the same for order-level coupons in the Order Coupon report. The Product Coupon report shows you how effective product-level coupons are in terms of revenue, unique purchases and product revenue per purchase. If you’re using internal promotions, for example internal banners that promote sales on another part of your site, you can track views, clicks and the click-through rate for those promotions in the Internal Promotion report.
What is a multi-channel funnel report?
These enable you to see how all your channels work together to create conversions. In Analytics, conversions and e-commerce transactions are credited to the last campaign, search or ad that referred the user when they converted. But what role did prior website referrals, searches and ads play in that conversion? How much time passed between the user’s initial interest and their purchase? How many times did they return to the site before converting? The Multi-Channel Funnels reports answer these questions – and others – by showing how your marketing channels (that is, sources of traffic to the site, as covered in your Acquisitions reports) contributed to create conversions.
The Multi-Channel Funnels reports are generated from the conversion paths that led up to each conversion and transaction – the concept is the same as in the Goal Flow report we’ve already seen. By default, only touchpoints within the last 30 days are included, but you can adjust this to anything between one and 90 days using the Lookback Window selector at the top of each report.
Conversion path data includes interactions with virtually all digital channels, including: paid and organic search (on all search engines, along with the specific keywords searched), referral sites, affiliates, social networks, email newsletters and custom campaigns that you’ve created. Google Analytics automatically detects many of the channels that send traffic to your site, including unpaid search (all search engines), referrals from other websites (including social media sites) and direct traffic (users typing in your URL or bookmarking it); other channels require some setup in order to be tracked, including AdWords, paid search on non-Google search engines, and custom campaigns.
A channel may be the First Interaction in a conversion path, Last Interaction or, if it is neither of these, an Assist Interaction. The reports give you a count (and a monetary value) for each of these and all combinations, so you can work out how important any given channel is in initiating, assisting or completing conversions.
The Overview report here contains a “Multi-channel Funnel Visualizer” to give you a visual representation of how different channels are working together to create conversions.
The Assisted Conversions report shows how many sales and conversions each channel initiated, assisted and completed, along with the value of those conversions and sales.
The Top Conversion Paths report shows the paths that your customers took on their way to purchase, plus the number of conversions from each path and the value of those conversions. This allows you to focus on the flow rather than the contribution of each channel.
The Time Lag and Path Length reports show how long (in days and in touchpoints respectively) it took for users to ultimately become customers.
What is an attribution model in Analytics?
Under Attribution you can refine how credit is assigned to the various channels that might have contributed to a conversion.
An attribution model is a rule, or set of rules, that determines how credit for sales and conversions is assigned to touchpoints in conversion paths. For example, the Last Interaction model in Google Analytics assigns 100% credit to the final touchpoints (i.e., clicks) that immediately precede sales or conversions.
In contrast, the First Interaction model assigns 100% credit to touchpoints that initiate conversion paths. In the Linear attribution model, each touchpoint in the conversion path would share equal credit for the sale. In the Time Decay attribution model, the touchpoints closest in time to the sale or conversion get most of the credit. In the Position Based attribution model, 40% credit is assigned to each the first and last interaction, and the remaining 20% credit is distributed evenly to the middle interactions.
Under Attribution you can choose between Google Analytics’ default attribution models or even create a custom model using the Model Comparison Tool. Do read Google’s discussion of some of the considerations involved.
Note that conversions are counted differently in Analytics and AdWords, and be aware that some experts see shortcomings in both.
3.7 Customization tab: An Overview
What is the customization tab in Google Analytics?
Under the Customization tab you can create custom reports, access custom reports you’ve previously created (if any) and, if you have a Premium account, access unsampled reports.
Custom reports are those you create to track something Google Analytics doesn’t already track or to present data in a specific way. You pick the dimensions (City and Browser, for example) and metrics (such as Visits, Pageviews and Bounce Rate), and decide how they should be displayed.
How to create a custom report in Google Analytics
1. Create custom report
Click “+New Custom Report” at the top of the table under the Customization tab (if this doesn’t appear, first click Overview under Custom Reports in the navigation sidebar).
2. Set title and number of report tabs
Enter a title, then optionally click “+add a report tab” if you want the report to have more than one. Note you’ll need to repeat the following steps in turn for all the other tabs you create.
3. Select a report type
Choose between: Explorer (the usual Analytics report, with a line graph and data table), Flat Table (just the data table) or Map Overlay.
4. Define your dimensions and metrics
See the main text, below.
5. Add an optional filter
If you want to display only a subset of the dimensions – for example, if Browser is a dimension, you could choose to display only data relating to certain specified browsers – then define a filter for it.
6. Restrict views
You can also optionally select whether this report should appear only in selected Views or in all Views associated with this Account. Finally, click Save.
To build a custom report, you’ll need to specify at least one dimension and one metric. Dimensions are characteristics or attributes that describe something, such as Visitor Type (new or returning), Source (the name of a referring website or search engine), Product name or code, and so on. Metrics are individual elements of a dimension that can be measured as a sum or a ratio – so as a rule of thumb, things you count.
For example: “There were 350 Pageviews and 15 Transactions from new visitors, and 967 Pageviews and 40 Transactions from returning visitors.”
Pageviews and Transactions are the metrics and Visitor Type (new or returning) is the dimension.
Note that some metrics and dimensions can’t be paired in a custom report, and Google has a guide to Valid Dimension-Metric Combinations.
How to create a custom dimension
It’s possible to create a custom dimension not available in Analytics by default. To do this, click Admin, select the appropriate Account if you have more than one, and select the relevant Property. Click “Custom Definitions” in the Property column, then choose “Custom Dimensions”. If the dimension you want is not already in the list, click “New Custom Dimension”. You might wish to do this, for example, to integrate a third-party app with Analytics to import data from it, but note that defining the dimension might require some technical understanding.
Once you’ve created custom reports, you can access these under the Customization tab – they’ll be listed under Custom Reports in the navigation sidebar on the left. (If they don’t appear, first click Overview under Custom Reports.) Click a report name to view it; while you’re viewing it, you can click Edit in the menu bar to make changes to it. To delete an existing custom report or organize them into groups, click Overview and select the option you want from the Actions drop-down menu.Read previous article Read next article
3.8 Admin tab: An Overview
What is the admin tab?
Click Admin at the top of any Analytics page to switch to the tab where you manage Google Analytics accounts, Properties, Views and users. We’ve already seen a few examples of things you can do here. Use the menu at the top of each column in turn to select the account, Property and View you want, or to create new ones. Click the appropriate entry in each column to access the Analytics page for the activity you want – for example to manage users, create new Properties and Views, create data filters, whatever it may be.
Analytics keeps track of what’s done in the Admin tab. In the Account column, if you have more than one account, select the account you want, then click “Change History” to view what has been changed in that account over the last 180 days, by whom and when.
Bear in mind that to use the options in the Admin tab, your account requires Edit permissions. If you set up the Analytics account and are the account administrator, you’ll have this automatically; if you create new users, you can give them different levels of access.
3.9 Segments: An overview
What are segments in Google Analytics?
Segments are a way to pull out and analyses subsets of your data. If, for example, you want to look at UK males aged 18-24 who came to your site from Facebook, you can construct a Segment that applies this combination of criteria, and then view the data about this specific group in any report you wish.
Two things to note about this: first, Segments are non-destructive filters that do not change your underlying data – they are just a way of pulling out the data relevant to what you want to look at. Second, once you apply a Segment, it remains available as you navigate throughout the reports until you remove it, which means you can access all the relevant data in all the reports you’d normally use. In other words, Segments mean you don’t have to focus on your entire audience, your whole site or the complete conversion flow; you can focus in on any part of this selectively. What’s more, you can have up to four Segments active at a time, and compare results from each Segment side by side in your reports.
Google Analytics includes predefined Segments (“system Segments”) such as “Paid Traffic” and “Visits with Conversions”, which you can use as-is or duplicate and edit to create new custom Segments. You can also build your own Segments from scratch, share your custom Segments and even import Segments created by other Analytics users from the Analytics Solutions Gallery: click Segments in the left-hand panel to see what’s on offer.
There are a few limitations to bear in mind:
- There’s a maximum of 1,000 Segments per Analytics account.
- Each View can have up to 100 Segments.
- The date range maximum when using Segments is 90 days.
- Don’t use Segments with Multi-Channel Funnel reports (see the relevant section above).
Because Segments don’t change your data or the way it’s collected, you can apply a Segment to data collected before you created it – remember, it’s just a way of pulling out data you want to look at. Bear in mind, though, that Goals, Events and other user-defined types of data are recorded only from the point when you define them onwards, so if you create a Segment that uses such user-defined data types, the Segment will have no access to data from before that point.
How to use segments in Analytics
To apply Segments in a report, open the View that contains the reports you want, switch to the Reporting tab and select the report. At the top of the report, click “+Add Segment”. You’ll see a list of all available Segments, including the predefined system Segments and any Segments you’ve previously created or imported. Tick the Segment or Segments you want to use (up to four at a time), then click Apply. You can now switch to any other report if you wish; the Segment will remain active until you remove it.
To quickly remove a Segment, look at the top of a report to see which Segments are applied, mouse over the relevant Segment to open its menu, then click Remove. To duplicate it, do the same but select Copy in the menu; the copy opens in the Segment Builder, where you can then modify it – for example by clicking “+Add Filter” to add an additional condition to the Segment definition. When done, click Save.
The menu includes a third option: Remarket enables you to create a remarketing list based on the Segment.
How to create a segment in Google Analytics:
1. To begin, open any report and click “+Add Segment”.
2. In the Segment list that opens, click “+New Segment” to open the Segment Builder.
3. Enter a name for your new Segment.
4. To configure the criteria for the Segment, start by clicking a category of dimensions or metrics in the left-hand panel – Demographics, say – and use the options that appear in the right-hand panel to define the filters you want to apply to your data.
5. Where relevant, use the pop-up menus to select operators such as “Contains”, “Exactly matches” or “Starts with”, and type in any specific strings you want to match.
It’s possible to create quite complex configurations of criteria, but the logic of how they combine can be a bit confusing at first glance. In brief, if you apply multiple values within the same dimension, then by default OR logic applies – so for example data will be included if it matches either of the criteria Age 18-24 OR Age 25-34. Otherwise AND logic applies, and data will be included if it matches both criteria – for example, “Demographics/Age 18-24” AND “Demographics/Gender: Female”. To create something different, see the “Advanced options” section below.
In many cases, you’ll also need to specify a Scope for the data you want:
- Hit: behavior confined to a single action – for example, viewing a page or starting a video.
- Session: behavior within a single session – for example, the amount of revenue that users generated during a session;
- User: behavior across all the sessions within the date range you’re using, up to 90 days – for example, all the goals that users completed or all the revenue that they generated (across all the sessions) during the date range.
Click more dimensions in the left-hand panel to add further criteria and define the Segment more precisely. When you’ve finished, click Preview to apply the Segment to the current report (don’t worry about Test). If this doesn’t show you the information you expected, you can continue editing the Segment, then preview again. When you’re happy with it, click Save to close the Segment Builder, save the Segment and activate it for all your reports.
You can also use an existing Segment as the starting point. For example, you could start with the system Segment “Sessions with Conversions”, duplicate this and then add additional filters for criteria like Country/Territory and Campaign to focus on data about specific countries where sessions with conversions originated, and data about specific campaigns that led to sessions with conversions.
If the Segment you want to start with is already applied, start as described under “Using Segments” above; otherwise, open the Segments list as just described, locate the Segment you want to start with, click Actions next to it, and select Copy. A duplicate of the existing configuration opens in the Segment Builder; to avoid confusion, give this a new name straight away, then edit the existing filters or add new ones (or both) as required. When done, preview as above. When you’re happy, click Save. The original Segment you used as the starting point will remain unaffected.
To import a Segment, open the Segments list. This time, click “Import from Gallery” to open the Analytics Solutions Gallery and go straight to the Segments category. Use the sort and filter options to further narrow the content. When you find a Segment you want, click Import below the Segment description. You can opt to make the Segment available in all Views in the account you’re currently using, or choose a single View. Click Create and you can then rename, modify and configure the Segment as required, and finally Preview and Save it as usual.
What advanced options are available in Google Analytics?
In the Segment Builder, under Advanced in the left-hand panel, you’ll see two options: Conditions and Sequences. These enable you to build up more complex sets of criteria based on dimensions and metrics as we’ve seen, but with some additional options:
- You’re not restricted to specific categories: these let you create filters for any dimension or metric.
- Include or exclude specific data: they can also include both AND and OR conditions.
- When you include user and session-based rules in the same filter, those are joined with AND logic – that is, data is included when it meets both conditions.
- Sequences filters: these let you determine whether the sequence begins with the first user interaction or with any user interaction.
- When you include multiple steps in a Sequences filter, you can specify that one step follows another at any time or immediately. The subsequent step can occur in the same session or in a subsequent session.
The Segments you create can be as straightforward or as complex as you choose. You could, for example, look at all the users who first visited your site in a specific month, or even on a specific date, and trace exactly what they’ve done since then. (Just to complicate matters, Google uses the name “cohort” for a Segment that includes some type of date condition.) Or you could compare users who convert with those who don’t, and see if you can identify any patterns in their behavior that might help you turn more of the latter into the former. Making the most of your Analytics account can often require one of the many Analytics courses online available to ensure your knowledge is up to scratch!
You could look for “high-value users”, such as those who purchase or visit frequently, have done so recently, and account for more than some specified value – say £100 per user. From here you could then navigate through your reports to find out more about these users – their location, demographics, technologies or channels used, for example. Armed with these insights, you can then develop your audiences and marketing around that data.
One final scenario: you could create a Segment of users who viewed product-detail pages, clicked Add To Cart, but never actually placed orders – on the face of it, these users have indicated a strong interest in purchasing, so it should be worthwhile to try enticing them back with a remarketing campaign. Google has a page that shows you exactly how to configure the criteria for such Segments, helping you master how it’s done.Read previous article Read next article
3.10 Views and filters: An overview
What are views and filters?
We’ve seen that within your Google Analytics account you can set up one or more Properties, each of which tracks at least one website or app. Within each Property you can create multiple Views. Google used to call these “Profiles”, but the term “Views” does give you a hint as to how they work: a View is a defined perspective – a window, if you like – on the data related to that Property. It might be a panoramic picture window, so to speak, which gives you the complete picture; and in fact, one default unfiltered View is automatically created when you set up the Property, giving you access to all the data for that Property.
You can, however, create additional Views that give you a more restricted picture by defining filters, which Analytics deploys to exclude some data, include some or transform the raw data it is receiving.
Why would you want to do that? One common example is to give you a truer picture of traffic to your site by excluding visits made by yourself or your staff. Instead of counting every hit, you’d set up a filter to exclude traffic coming from IP addresses within your business, so that what Analytics shows you in its reports is just genuine visitor traffic, which is – of course – what you really want to know about.
In the same way, in addition to the default View of all the data for your site, you could create a View of the traffic to a subdomain such as sales.mysite.com, so that you can see the data relevant to that in all the available reports within that View. Or you might wish to have a View of just the AdWords traffic to your site. If you send both web and app data to the same Property, you might want to create specific Views that enable you to analyses just web data or just app data separately, in addition to the View or Views that track both together. In each case, for each additional View you create, you apply filters to them so that they each include the specific subset of data you’re interested in.
A few important points to note about Views in Google Analytics:
- All incoming data related to the Property gets sent to all Views within it: however, Analytics applies the filters defined for each View to process the incoming data and this processed data is what you see in all the reports within each View. Applying a filter means Analytics will ignore or throw out all data that doesn’t match, and that excluded data is simply not available in any reports within that View. For all practical purposes, it just doesn’t exist. This is the crucial difference between applying filters at the View level and using Segments.
- The filters you’ve defined will apply to all reports within that View: this means that some reports might have no content, if the data that they’re designed to report on has been filtered out.
- Processing by other Analytics reporting tools, such as Goals, Segments and Alerts, is applied within individual Views: you’ll recall that when you set up any of these, you first must select the Property and then the View in which you want them to operate. All these tools can work only on the data in that View, which has been pre-filtered before they receive it.
- When you create a View, Analytics can start reporting on that specific data from that date forward: reports in that View will not contain any data collected prior to the creation date of the View. If you need to view reports dealing with data from before that date, then you can use the original unfiltered View, created automatically when you first set up the Property, and use the date range and other controls to isolate specific information. Be careful, however, not to apply filters to that original View or you’ll lose all access to the unfiltered data.
- In the same way, if you delete a View, that specific perspective of the data is gone: So, don’t delete a View if you think you might ever want to report on that specific perspective of the data.
For these reasons, Google strongly recommends that you don’t delete the original, unfiltered default View or add filters to it. When you add filters, the data they exclude becomes “unavailable”, permanently. When you delete a View, that specific historical perspective of the data is gone, forever. So, if you want to create a filtered View of your data while preserving all the original data, create a copy of the original, unfiltered View or set up new, additional Views and then customize each one to meet your specific reporting needs in each case.
You can add up to 50 Views to a Property – remember, you need to have an account with Edit permissions (or be the account administrator) to add Views:
How to create a new view in Google Analytics:
1. Select Property
Click the Admin tab and select the relevant Account if you have more than one. Next, click the Property to which you want to add the View.
2. Create new View
In the View column, click the drop-down menu and select “Create new View”.
3. Choose and name View
Select either Website or App as appropriate (see below for more), and then enter a name for the View – use something specific and descriptive of the data you’ll be filtering for, so you’ll be able to easily pick this View from a list in future.
4. Pick time zone
Next, select the Reporting Time Zone (see section 4.1 of the ‘Google Analytics: Getting Started’ playbook for more on this). If your Analytics account is linked to a Google AdWords account, the time zone is automatically set to your AdWords preference and you won’t see this option.
5. Create View
For a standard reporting View, leave User ID OFF (see below), then click Create View.
After you create a View, you can come back to the Admin page and edit the View settings.
If your Analytics account is linked to an AdWords account (see chapter 2.5 of the ‘Google Analytics for SEO, SEM, Website and CRO’ playbook), data from the AdWords account is automatically imported into any new View you create on that account before – of course – being subjected to any filters you’ve set up on the new View.
What is the difference between a web or app view?
When you create a View, you can choose between an App View and a Web View. Google says these two View types “give you a slightly different analysis experience but are otherwise the same”. For example, App Views include some reports that aren’t available in Web Views, such as Crashes and Exceptions and the Google Play reports, while Web Views give you Site Content reports. It would seem logical to choose according to the type of Property you’re dealing with, or the more important type if you’re including both websites and apps in the same Property and want to report on them together in the same View.
Create a copy of a view
Google warns that some Analytics features – notably filters – fundamentally alter how data is collected or processed in your account in ways that cannot be reversed. As a result, you should always duplicate the original View before making changes. Always keep the original View unchanged and add filters or other reporting features to the duplicate Views to meet your specific needs. (You have to wonder why Google doesn’t just lock the default unfiltered View and make it impossible to edit or delete…)
Duplicating a View is a straightforward process:
How to create a copy of a view in Google Analytics:
1. Click the Admin tab, select the relevant Account and Property, and navigate to the View you want to copy.
2. In the View column, click “View Settings”, then click “Copy View”.
3. Give the new View a distinct name, so you can distinguish it from the original, and click Copy View to confirm. Job done!
When you duplicate a View, settings and features controlled at the View level (like filters, Goals, users and their permissions) are duplicated in copied Views. Cost source links and shared assets (like annotations, Segments and alerts) are not duplicated into copied Views.
How to edit view settings
Assuming you have Edit permission, you can change a View’s settings at any time. Click the Admin tab, select the relevant Account and Property, and navigate to the View you want. In the View column, click View Settings.
You can change the following elements – when complete, click Save to enable your tweaks:
- The View name: perfect if the name you gave your View doesn’t make sense.
- The Website name used by Content reports: including In-Page Tracking (but note that this does not change the actual URL being tracked – that’s set at the Property level and depends on the unique identifier embedded in the site code).
- Reporting Time Zone: this determines when Analytics regards each day as beginning and ending for reporting purposes (see section 4.1 of the ‘Getting Started with Google Analytics’ playbook). Note that this applies from when you make the change, not retrospectively, so there may be a glitch in your reports at the transition point.
- The default homepage for your site: usually index.html or default.html. Changing this affects how page information appears in your reports. If you’re not sure, leave this blank.
- Exclude URL Query Parameters: any query parameters or unique session IDs (for example sessionid or vid) that appear in your URLs but you do not want to see in your reports.
- Currency to display: self-explanatory.
- Bot Filtering: select this option to exclude sessions from known bots and spiders.
- Site Search tracking: this must be set up separately for each View – see the Site Search section, above.
Understanding how to use filters
The View settings are useful, but of more importance are the filters you apply to a View. As we’ve noted, Analytics will apply the specified filters to exclude, include or transform data before it’s reported on in each View, so it’s vital to be certain of what you’re setting them to do, or the data you see could be incomplete or misleading in unexpected ways.
Analytics offers a range of predefined filters – examples include excluding or only including traffic from a specific domain, such as your ISP or company network. You could also opt to exclude or only include traffic from specified IP addresses, or to exclude or include only traffic to specified subdirectories, among others.
You can also define custom filters, which may be Exclude or Include filters, Search and Replace, convert Lowercase to Uppercase or vice-versa (handy where some reporting tool is case-sensitive). There’s also an Advanced filters option that combines multiple fields.
It’s vital to understand how filters operate, because of their fundamental effect on the data you end up seeing. For example, setting up an Exclude filter to exclude the Chrome browser will exclude all information about anyone who visits your website using Chrome, so you’ll see no user, path, referral or domain data for these. Google has a comprehensive reference guide to filter types and uses, which you should make a point of studying before you get stuck in.
Broadly speaking, however, filters work in the standard way according to ordinary IF-THEN logic. You begin by selecting the Filter Type (Predefined or Custom), then the Filter Field – the type of data you want to evaluate or change. This may be User IP address, device type, geographic location, and so on. You then specify a Condition and from this pick an Action that tells Analytics what to do if the condition is true.
An example would be to create a simple Include filter that specifies if Country exactly matches UK, then include that data in your view. This would mean that all data from countries other than the UK would therefore be ignored.
If you’re using predefined filters to do things like exclude traffic from specified IP addresses, or IP addresses beginning with a certain string, then you simply specify your business’s internal IP addresses using “Regular Expressions” (which for IP addresses means a format like this: 163\.212\.171\.123 – Google has a guide to the details).
Bear in mind that filters, like all configuration settings, are not retroactive; as we’ve noted, they apply from when you create them onwards. This means there will be no data available from before you created the filter, and you need to watch out for gaps or glitches in your data if you change a filter.
Note also that filters are applied in the order they’re listed in your configuration – and order matters because the output from one filter becomes the input for the next filter in the chain. If, for example, you want to measure traffic from North America, you need to create an Include filter to include data where the Country is the US OR Canada. If you create one filter to include US only and then another to include Canada only, you’ll end up with no data: the first filter will output only US users, so the second filter will find no Canadians and output zero data.
For this reason – and because filters fundamentally change what data you’ll see – Google emphasizes that when you’re setting up a new filter you should always test it on your account’s Test View first. Only when you’re certain it does what you expect it to should you save and apply it. When you do save a new filter, it’s added to the Filter Gallery for your account and can then be applied to any View within it.
Creating and managing view filters
It is possible to create a filter at the Account level and apply it to multiple Views, but we’ll focus here on creating a filter for one View.
How to add and apply a view filter in Google Analytics:
1. First steps
Click the Admin tab, select the relevant Account if you have more than one, then navigate to the Property and View you want. In the View column, click Filters, then “+New Filter”.
2. Set up new filter
Select “Create new Filter” and enter a name for the filter.
3. Choose and define filter
Select Predefined filter and choose the filter you want, or select Custom filter and configure the options following our guide above.
4. Preview and save
Before you commit to using the filter, click the “Verify this filter” link to see how it will affect your data reporting. When you’re happy with the filter, click Save to enable it.
5. Apply to other Views
To apply an existing filter to another view, first navigate to that view, then click “+New Filter”. This time, choose “Apply existing Filter”, then select the filter or filters you want to use with that view before clicking Add>> to apply them. Click Save to finish.
Understanding User ID views
We’ve seen that when setting up a View, you get an option to switch on User ID View. This applies only to User ID enabled Properties. A User ID View is a special reporting View that displays only data from sessions in which you send unique ID and related data to Google Analytics. This View enables you to analyses the segment of traffic with an assigned ID separately from your other traffic.
What does this mean? User ID is essentially a way of identifying unique users across multiple devices and different sessions. This gives you a more accurate user count and a more rounded picture of engagement with your business or brand, but it requires you to be able to generate your own unique IDs, consistently assign IDs to users, and include these IDs wherever you send data to Google Analytics. This usually means you need to have a sign-in and authentication system, with robust data protection. Then, you could send the unique IDs generated by your own authentication system to Google Analytics as values for the User ID. Any engagement that takes place while a unique ID is assigned, such as link clicks and page or screen navigation, can be sent and connected in Google Analytics via the User ID.
User ID Views include a set of Cross Device reports, which aren’t available in other reporting Views. These reports give you the tools you need to analyses how users engage with your content on different devices over the course of multiple sessions. Bear in mind that the Property must be User ID enabled, you need to set up the User ID integration in your Analytics tracking code, and you need to add a new View to the Property and specifically make it a User ID View. Note also that User ID Views display only data from sessions in which a User ID is assigned and related data is sent to Google Analytics. They won’t include any data from sessions without User IDs, so you’ll need to have different Views in addition to see and analyses any every day, non-authenticated traffic, including any new users who haven’t (yet) signed up.Read previous article Read next article
3.11 Report Annotations: An Overview
What are report annotations?
Annotations are short notes that you and other users can write and attach to specific dates in your Google Analytics reports. Even the most basic level of user can create annotations so anyone who can access a View can annotate it.
Click the tab below any time-based graph (immediately beneath the graph’s time axis) to open the Annotations drawer and both see annotations that have already been entered and add your own. Type your note into the text area, and optionally click the Private button next to Visibility if you don’t want others to see it; by default, annotations are Shared, which means that any user with access to that report can view it. Annotations you create are automatically attributed to the name you logged into Analytics with.
You can use annotations to call attention to something of interest, create a reminder or to-do, or add explanatory information that helps to shed light on what’s recorded in the graph – noting that a spike in traffic correlates with the launch of a new marketing campaign, say, or recording that a dip in the figures coincided with a site outage. There are no restrictions on what you can enter – it could be general news, weather or any other time-specific factor that could affect website behavior.
After you’ve created an annotation, a small caption icon appears in the graph’s time axis below the corresponding date. You can click this icon (or the small arrow tab again) to display that annotation, as well as any others for the selected period.
Annotations are replicated among all reports within the same View. For example, if you create an annotation in the Landing Pages report, the caption icon will also appear in the All Referrals report. You can also view all annotations (shared or private) in a list in the Admin tab for that View, assuming you have admin rights.
4.1 Audience Segmentation: An Overview
What is audience segmentation?
Marketing and advertising commonly concentrate on reaching the largest possible audience, but this scattershot approach is increasingly seen as wasteful compared with more targeted strategies: online conversion rates remain as low as two to three percent, compared to in-store rates of 20 percent or more. It makes sense, therefore, to take a more focused approach, aiming to target the right customer with the right product at the right time, based on an informed idea of what that individual’s motivations and needs might be. Pitching your marketing activity at a generalized audience or even a notional “average” customer just won’t cut it.
Of course, you’ll rarely know all about your individual users. Modern data capture techniques make it possible to address remarketing emails personally to an existing customer, which hugely increases click-through rates and consequently the odds of a successful conversion. You’ll still need to spread the net more widely to grow your business, however. The solution lies in the fact that large customer bases can be divided along clearly definable lines into “segments”, which can then be targeted more specifically than the entire audience. Market segmentation gives you the opportunity to more efficiently deploy resources and develop messaging that resonates with specific segments.
Before you decide to go down the road of audience segmentation, there are a few basic criteria to consider, which will necessitate some initial investment in research or even undertaking a class with Imparture!
What criteria should be considered when defining Google Analytics segmentation?
- Market size: your market must be large enough to justify segmenting.
- Difference: there must be a measurable difference between the segments.
- Accessibility: the segments you determine must be logistically reachable.
- Profitability: you need to be confident that there will be a projected net profit after the additional costs of segmented marketing.
- All market segmentation needs to be based on solid user data. you can expect better results, however, if your divisions are more nuanced and more highly tailored to the psychological and emotional characteristics of your users, not simply their location and gender.
4.2 Segmentation categories
What are the segmentation options in Google Analytics?
Breaking down your audience by gender and age range is still a good place to start. From the Google Analytics Demographics report you should be able to see what age group is most frequently visiting your site, and also – importantly – its conversion rates. Create a segment for this age group and look at the content report: what pages are they viewing? What sources are they coming from? And what landing pages are these sources taking them to?
How closely do your demographic conversion numbers match your target customer profile? Are your pages and imagery appealing to the ideal customer? Could you adjust them to appeal more to your target customer or to the users who currently bring in the most revenue, or is there an opportunity to appeal more to your poorly performing segments? Is there enough revenue opportunity to make the necessary changes to your site, or invest in targeted email marketing, paid search or remarketing?
Be warned, though, that some of the demographic details you get could be largely guesswork. Google uses a variety of methods to ascertain the age of visitors, ranging from simply grabbing the age from the user’s Google or social media profile, to extrapolating an age based on the websites that the user frequently visits, specific URLs they visit (such as: https://www.imparture.com/course/online-training/google-analytics-online/). For instance, if you visit a lot of female shopping sites, are often active on social media and commonly read fashion blogs, then Google might assume you’re a female in your late teens to mid-20s.
The next step beyond demographic audience data is psychographic, which looks not just at who your audience are, but at what they believe in and what motivates them. While demographics give you the “who,” psychographics give you the “why.” If you’re selling outdoor equipment, for example, it’s not enough to know that a potential customer is a woman in her 30s: you need to know about her lifestyle and interests – someone whose only interest is cooking will be far less likely to purchase than someone who goes jogging or runs marathons.
You might need to draw data from all kinds of sources to build up a rounded picture: survey your customers, and look at their social media profiles to find out what makes them tick. Meet them where they are – if you’re concentrating on marketing on Facebook, but your customers spend more time on Pinterest, then you need to think about changing your game plan. You should even look at what kind of ad gets more click-through: if it’s emotionally-driven rather than factual, then you have a clue to what appeals to your users. You need to appeal to both their needs and their desires, and offer irresistible deals. By understanding your users’ personality drivers and the motivations behind their purchases, you can make sure you’re using the right channels and the right messages for the right audience.
This should extend to stepping back from your marketing and advertising materials and seeing them through your audience’s eyes. Understand your prospects’ “perception filters”: each word and image you put on your website is colored in their minds by their own unique psychographics, demographics, location, cultural biases, and experiences with your brand and with other brands. Does the photo of a smiling person on your home page convey warmth and sincerity, or does it look artificial, like a stock photo? If you want to attract customers from China, say, have you got an appropriate ethnic mix of photos on the site? Don’t get caught up in endless second-guessing, but be aware that different people can interpret the same images and words in different ways. The key is to have a clear picture of how you can expect things to be perceived by your critical psychographic segments.
Google Analytics gives you a very direct way to build up a picture of your users’ interests. As they visit sites that are part of the Google partner network, Google associates visitors with certain category groups based on the type of sites they frequently visit. Analytics’ Affinity Categories report tells you about these on the broad level (“Travel Bug”, say), and the Other Categories report on a more specific level. Bear in mind that this data is not available for every single user, so the reports are based on a percentage of your visitors.
Take a look at visitors coming to your site, specifically in the light of their specific interests as revealed in Other Categories, and look at how well or how poorly they’re converting. Where are they coming to your site from? Are they using a specific source or medium to access the site? If your key visitors are coming from Twitter, for example, then it could make sense to become more active on Twitter. If they’re frequently being referred from the same site, could you be targeting your display advertising (or content) at that site or similar sites?
Another great use of this data is for remarketing, because it enables you to target specific groups of users that you know either convert well or are typically interested in your content. If, for instance, you find that new visitors aged 25-34 in the “Travel Bug” category don’t go on to buy, could you target this group with a unique discount to encourage them to return to the site and go on to convert? It’s remarkably straightforward to do this: simply create a new remarketing list in your Property settings and click “Create my own remarketing type using Visitor Segments”.
With the variety of platforms available today, it’s essential to know how many of your customers are using mobile devices such as smartphones or tablets. How many of these are converting? If tablet users convert far more than smartphone users, why might this be? Under Audience > Mobile > Overview, Analytics can show you the share of traffic between different device types, and the contribution of each to the total revenue generated. You can see how the mobile platforms compare with others (but make sure you pick a metric that gives you fair grounds for comparison, such as the e-commerce conversion rate or the average visit duration). Under Audience > Mobile > Devices you can break down the reports into individual devices, brands or even different screen resolutions.
This level of detail can provide very useful clues to how to improve your site’s performance. For example, if conversions are poor on smaller-screen devices, look at your site design. Is a key page element “below the fold” on smaller screens? If this specific element is vital to the success of the page, are there cues to help ensure that users find it? Remember that mobile devices are used upright as often as sideways, so look at what happens to your page if a device is held in portrait rather than landscape orientation: does something important, such as an action button, fall too far down the page for users to bother scrolling down to?
Also consider the advertising on your pages. Experience shows that a single persistent advert across the top of the web page that remains permanently in view will perform well on a smartphone. On a tablet, skyscraper and banner adverts might work more effectively because a persistent ad is considered annoying. Use your Analytics information to assess how different ads perform based on the layout and screen size.Read previous article Read next article
4.3 Audience Personas: An Overview
How to identify audience personas in Google Analytics
In addition to identifying broad audience segments, it’s a useful exercise to formulate what the marketing industry calls user personas. A persona is a step beyond a description of your notional target customer or segment; it’s a more rounded fictional character who encapsulates and represents some distinct part of your audience. The advantage of developing personas is twofold: first, thinking through the characteristics of your personas helps you identify important aspects of your actual or target audience more clearly. And second, holding a user persona in mind as if it were a real person helps you personalize your site, apps and communications, making it all more targeted and more engaging.
So, what does a persona cover? A fully fleshed out buyer persona includes everything from demographic information to hobbies, and from career history to family size, all written as if the persona were a real person. It should provide insights into your customers and what’s relevant to them, including what they’re looking for, how they go about finding it and what factors might influence them to make or break the purchase.
What to consider when defining personas in Analytics?
- How do they use social media?
- How much time do they spend online?
- Where do they work? What are the biggest challenges they face at work?
- Which blogs, news sources or media do they consume on a regular basis?
- What are their communication preferences?
- How do they find their information?
- What is their previous customer experience and what do they want to change about it?
- What gets them out of bed and what keeps them up at night?
How to create audience personas
How do you develop user personas? Many businesses begin with traditional market research methods, including questionnaires, and look for patterns in the characteristics of their actual customers. Talking to your customer-facing staff can be invaluable. There are some very helpful, well-established templates available online which you can use to organize your data and your thinking.
You can also draw data from your CRM, consumer research and social channels. You might decide to conduct focus groups or interviews to gain further insight, in which case you should approach both existing and potential customers.
Within Analytics, you can create four broad personas simply by combining “site average pages per session” with “site average session duration” or time on site. This should give you enough data to create four segments:
- Methodical Mary: visits the site for a long time and looks at more pages than average. These users are taking their time to decide.
- One-hit Juan: visits few pages but looks for a long time. Might be enjoying your video content, or knows exactly what they’re looking for.
- Lost Lucy: hits a lot of pages in a short time but doesn’t spend any significant amount of time on any. May be looking for something but not finding it.
- Bouncy Bob: visits fewer pages than average and stays for a shorter time. Possibly just a casual browser.
Once you have these segments, you can examine each one’s purchase rates, page flows and devices used. For a more detailed picture, you can begin by looking at the keywords that brought users to your site:
1. Start looking
Under Acquisition > All Traffic > Source/Medium, set the Primary Dimension to Keyword. This will show keyword searches. The vast majority will be “not provided”, but the results you’ll get might be enough to get started.
2. Organize keywords
Now look at the keywords and group them into categories or themes. Use this to identify people searching for these themes. Make a list of personas such as “Serious runners interested in buying sneakers locally for under $100,” or “Beginners thinking about getting into mountain climbing.”
3. Refine by referrals
Under Acquisition > All Traffic > Referrals, click the Secondary Dimension drop-down and select Landing Page. The patterns and preferences you find will help you create a sharpened buyer persona for each social network on which your brand has a presence.
4. Fill out the detail with social media
On Twitter, for example, look at “your followers also follow”. On Facebook, you can use Graph Search to learn more about the interests of your fans by running queries such as “pages liked by fans of [your business page]”. Examine the smaller social media sites your audience frequents to narrow down their interests.
How to use audience personas
How do you use customer personas? Once you’ve created a persona, you can adjust everything from the words you use on the phone to the content that’s served up on your website to ensure that prospective buyers receive the sales pitches that will be most persuasive to their personal situations. In traditional face-to-face or telephone marketing, you might ask a clarifying question (or series of questions) on initial contact to help you identify which persona is the closest fit, such as these:
- What do you see as your biggest business challenges?
- What one thing are you hoping to get out of a solution like ours?
- How do you see our solution helping solve your problems?
- Are you more concerned about [defining persona characteristic] or [characteristic of another buyer persona]?
- If I could help you with one thing, what would you like that to be?
Of course, the individual you’re dealing with might not perfectly match a predefined persona, or the answer you get might not be one you’ve anticipated. But an approach like this can be specifically helpful on your website. If you’re capturing leads on the site, include a clarifying question along these lines in your opt-in form, and you’ll have not just prospects’ names and contact information, but also an insight into how to appeal to them. In advertising or email campaigns, create separate ads or mailings using different language that will appeal to different personas, and include different response codes; that way, you’ll have a good idea of the likeliest relevant persona as soon as you receive each enquiry.
If you can deploy a responsive tool such as Hubspot, your site can serve up dynamic content based on the buyer persona you’ve assigned to your prospects: when your website detects their IP addresses during their visits, everything from alternative call-to-action buttons to entire blocks of content can be called up to appeal specifically to their interests.
You can also use Hubspot – or an alternative marketing automation program such as Marketo or Pardot – to tailor email campaigns to each prospect’s persona and stage in the sales process, or even do the same thing manually.
How to analyze your competitors’ audiences
There are tools available for more detailed analysis. A useful start is a free resource called Quantcast. Type a URL into Quantcast and you’ll get a whole range of demographic information about the site. To find out how visitors are responding to your competitor sites’ content, visit Quicksprout. Type in a URL, and Quicksprout will produce a list of the most shared pieces of content on that site. This will give you a good indication of what that audience is interested in, and therefore suggest what general content you might benefit from adding on your site or blog.
We’ve mentioned Google’s Industry Benchmarking reports in Analytics (Audience > Benchmarking), which enables you to compare your own sites with aggregated information from others in comparable industries. You can filter by location or devices used. But what if you want to compare against specific competitors? You can visit their sites and investigate them manually to assess what they’re doing or failing to do, and see for yourself whether there are any broken links or shortcomings in their site design and navigation; and you can also try a keyword search, using the terms you regard as critical, and find out how they rank compared to your site. All this, however, doesn’t give you the kind of detailed metrics to compare with what Analytics gives you for your own site.
There are many more. For example:
- Smart Insights: offers a huge roundup of tools for online competitor benchmarking.
- KISSmetrics: provides a blog post rounding up 25 online tools for finding information about competitors.
- InfoTrust: its blog lists six free tools for analyzing competitor site traffic.
5.1 Google Analytics for SEO and SEM: An Overview
How to use Google Analytics for SEO
Search Engine Optimization (SEO) has historically focused on maximizing a website’s ranking in search results by including the most appropriate keywords in the site’s content or meta data (such as tags or description). Ideally these keywords are the exact terms that actual or target users are looking for, which Google Analytics could help you glean because it reported data about the search terms they were using to reach your site. Analytics could even go a step further and give you an indication of how these terms correlated with conversions, so you gained a good idea of which terms were most likely to bring you in the best customers.
The catch is that an increasing number of search terms are reported as “not provided” in Analytics, because Google Search does not pass on the search terms entered in an organic search made using SSL.
This means you may have to work a little harder to find out something about the search terms that visitors are using, or rely a little less on keywords and take a broader approach to Search Engine Marketing, which may involve Pay Per Click and other strategies in addition to SEO. The good news is that Google Analytics is still hugely helpful to you in both approaches, especially when used in tandem with other resources such as AdWords and Google Webmaster Tools. This is especially true if you have taken the time to undergo a Google Analytics training course and can fully grasp all the available features.
This broader approach could actually pay additional dividends by freeing you from the tyranny of assessing your site’s effectiveness according to simplistic traffic measures. Instead, it prompts you to look at your users in the round, using the wide range of information about them that, as we’ve seen, Analytics has to offer. After all, even basic SEO shouldn’t be just a trick to get increased footfall on your homepage; it should be “really all about finding the right users using optimal content strategies.” You should be aiming “to learn more about your audience and provide meaningful content”, not hollow clickbait that leaves them disappointed with your site and won’t lead to conversions. Accordingly, instead of simple keyword data, “there are better ways to measure your organic search marketing performance, and many of them are already available for you within Google Analytics.” Equally, there are ways to optimize your site’s organic search performance without the level of organic keyword referral data you might have been accustomed to having.
5.2 Understanding Search Queries
How to understand search queries in GA?
Let’s start by looking at what analytical information you can glean from your audience using search queries, rather than simply finding out what keywords might be worth adding to your pages.
In a commercial context, one very important factor that search behavior can tell you about is user intent and, by inference from this, where the user is in the standard buying cycle. Many analysts divide search queries into three basic types:
These are searches performed to locate a specific website or service by name, for example “youtube”, “amazon” or your site’s name. Users performing this kind of search already know where they want to go. Navigational queries are common on mobile devices because it’s easier to enter a concise search term than a full URL, which would be counted as a direct visit rather than a search referral.
These users are most likely past the discovery phase of the typical inbound marketing sales funnel: they’ve heard of your brand one way or another, and might even be return customers. The intent isn’t necessarily to buy, though: they might be looking for information (or other content, such as YouTube video). That said, the fact that they search for the site rather than the information suggests that they’re quite confident of finding what they want on your site – they know your brand, and ostensibly trust and value it.
Transactional searches are performed to buy something, or with the intent to perform a specific action such as download a video game or find a local business. Examples include searches such as “buy PS4 online”, “pink iphone 6 case” and “Mexican restaurant in Tulsa”, but can also be the name of a specific – if generic rather than branded – product.
Not surprisingly, transactional searches produce the highest rate of conversions: users are likely to know what they want and be considering buying it, so your site’s success can depend on commercial factors such as price and availability. That said, not all product searches result in conversions, and some analysts who take a slightly different approach based on “commercial intent” have identified a class of user they characterize as the “tire kicker”, who may seem to be searching using these terms, but is unlikely to continue to a purchase (see below).
These searches are performed to find something out or answer questions – indeed, such searches are commonly framed in the form of a question: “how can I get rid of a stain” or “what golf clubs do the pros use”. This is the broadest type of query and tells us the least about purchasing intent. These users are most likely not ready to purchase: they are “high in the sales funnel and are just discovering they have a need for something”.
However, this doesn’t mean queries of this type won’t lead to a conversion: in the first example, they might be looking for a stain-removal product or discover that this is the type of product they need from the content on your site. These users might be in the research stage, at the very start of the typical purchase funnel, but if your site provides the information they’re looking for then this could both build value for your site or brand and also lead to conversions, whether these are actual purchases or sign-ups and the like.
The best way to exploit this kind of search, according to one analyst, is to identify specific search terms in your sector “that have high search volume and low competition,” if you can, “then get as much of that traffic as you can on an email list. That way, you’ll be the site on their mind as they’re ready to buy something.”
As we’ve seen, some analysts categorize search queries more explicitly according to “commercial intent” (that is, broadly speaking, intent to buy). In this model, there are four types of search query, ranging from high commercial intent down to low.
Types of informational queries:
- “Buy Now” search terms: typically used minutes before making a purchase. Examples include Buy, Coupon, Discount, Deal and Shipping, possibly associated with a product name (generic or branded). Such searches might not account for a large proportion of search volume, but they “convert like crazy”.
- “Product” searches focus on a product category (such as “running shoes”), specific product or service (“Macbook Pro”) or brand name (“Nike”), and/or terms such as Best, Top 10, Review, Comparison, Cheap and Affordable. People using such terms tend to be just a bit earlier in the buying cycle than people using “Buy Now” terms and accordingly convert just a little less. Interestingly, searches including words such as Cheap or Affordable “convert really well. For example, someone searching for ‘cheap laptops’ has already decided that they want a laptop… They’re just looking for a product in their price range.”
- Informational searches tend to include terms such as “How do I…”, “Ways to…”, “Best ways to…”, “I need to…” or just a task in the abstract (such as “removing carpet stains”). As you might imagine, people looking for information don’t tend to convert especially well, but this doesn’t mean this kind of search won’t assist in conversions later.
- “Tire Kicker” searches – characterized by terms such as Free, Torrent and sometimes Download – are very unlikely to convert, ever. It’s worth noting that specific product names can appear in any of these kinds of search: “a search like ‘watch The Simpsons online free’ is a classic Tire Kicker search. Good luck getting that person to buy anything (or even click on an ad). On the other hand, searches like ‘Buy Simpsons TV episodes’ (Buy Now Keyword), ‘Simpsons DVDs’ (Product Keyword) or ‘How to watch Simpsons episodes’ (Informational Keyword) will convert relatively well.”
According to Google Trends, informational queries – especially those including “how” and “what” – have increased exponentially in the past few years. A little over a decade ago, informational queries accounted for only about 48% of queries, but today almost 80% of queries are informational, with the remaining 20% divided between navigational and transactional queries. Marketing strategies haven’t always kept pace with this shift in user behavior: “marketers… are often caught up in the ‘need direct sales’ mentality by focusing too heavy on transactional queries and ignoring users who are still in the research and informational phase.” In the same way, analysts who use the commercial intent model also advocate focusing on “high commercial intent” searchers, who are on the point of buying. For example:
“Most SEO experts agree that – when it comes to choosing keywords — commercial intent is actually MORE important than search volume. [Users] stemming from informational searches are typically hard to convert into paying customers. Fortunately — with a little bit of research — you can easily find keywords that actual buyers use to search. And when you get your site in front of those people, turning them into leads and sales is a breeze.” 
Despite this, there are plenty of opportunities to convert business from each type of search query. If an increasing proportion of visitors are looking for information, this is what you need to provide: “it’s essential that you generate quality content to attract leads to your site and thus your product or service. The golden rule here: quality over quantity. You want to provide consistent, relevant, and reliable content that will help establish trust with consumers. Remember: they may not know what they want yet. It’s up to you to establish a relationship, gain their trust, and convert” by means of this more gradual and organic process.
Yet another way to categories search terms looks at how specific they are and accordingly classifies them as either “broad” (also called “head” and “short-tail”) or “long-tail” search queries. Broad search terms are short words or phrases that, while they may apply to your own industry and company, might also apply to every company in your industry or even to those in other industries (for example “shoes” or “loans”). Long-tail terms tend to be longer words or phrases that are more specific to your company or industry (such as “Nike red running shoes” or “commercial real estate loans”).
Broad search terms typically account for a high volume of searches, but there’s a great deal of competition to contend with. It’s difficult to rank high in such searches and win traffic from them, and the visitors you do get from such searches are less likely to become leads or purchasers. Because long-tail terms are more specific to you, however, the opposite applies to them: they represent lower search volume but little competition; it’s easy to rank in searches for them and win traffic; and visitors from these are more likely to convert.
These characteristics produce a remarkable result: while the most searched-for broad terms might bring you a large number of visits, you might easily get even more total visits – and almost certainly more conversions – from the range of long-tail terms combined. This statistical distribution phenomenon is what the term “long-tail” describes: the graph of visits shows a high number for the few “head” terms in your industry, but the numbers taper off very gradually for the range of “long-tail” terms, and adding up these many small numbers typically results in an even greater total than that for the head terms.
How long is the long tail? Hitwise puts it like this: if you had a monopoly over the top 1,000 search terms across all search engines (which is impossible), you’d still be missing out on 89.4% of all search traffic. There’s so much traffic in the tail it is hard to even comprehend. To illustrate: if search were represented by a tiny lizard with a one-inch head, the tail of that lizard would stretch for 221 miles.
In commercial terms, this means you really don’t need to focus on ranking high in searches for your industry’s head terms. “For highly-contested keywords in competitive markets, it could be more realistic and profitable to work on ranking well for the long tail. If you rank poorly for the head but rank well for much of the tail, you could still be very successful in business terms. It’s the bottom line you’re after, not the bragging rights that accompany ranking for your vanity term. You may lose one battle but still win the war.
“Once you accept that you ideally need to target multiple search phrases, the question is, how? Very simply, you need to come up with a variety of the phrases people will use in finding businesses like yours, and then create content (typically, pages or blog posts) focused on those terms.”
For SEO, this means it’s less critical to identify the exact “head” search terms, even if they are bringing a high volume of search to your site. It’s the bigger picture and the long game – or here, the long tail – that counts. In Google Analytics, “Instead of fixating on a handful of fat head keywords, it’s time to train executives to focus on what really matters to your business: how organic search brings revenue (and ultimately profit) into your organization.” So let’s look in Analytics for alternatives to measuring search performance based purely on specific keyword data.
5.3 Alternative measures in GA
Alternative ways to measure SEO with Google Analytics
Here are 10 ways to measure the value of your SEO efforts in a “not provided” world.
Measure the overall volume of organic search traffic over time
Instead of worrying about how your site ranks for certain keywords, focus on the bigger picture: overall traffic and conversions. Analytics can tell you about this in the Acquisition reports: in the All Traffic report, select Organic Search as the primary dimension and look at the pattern over a suitable date range such as the last year.
Segment organic search traffic by landing page
Still want some detail about keyword performance? In the Keywords report (which, remember, reports on all organic search traffic), select Landing Page as the primary dimension. This will give you a view of the pages on your site that are getting the most organic search traffic, which gives you a much quicker indication than individual keywords of which content is working for you. If the number one landing page on your site is the homepage by a large margin, don’t panic: this is common for established brands, and the homepage often represents a branded query. If you want to get advanced with your reporting, try inferring brand/non-brand to your landing pages based on the URL and reporting on each separately.
Use landing pages as a secondary dimension
Consider keeping Keyword as the primary dimension, and view Landing Page as a secondary dimension if there are still some useful specifics about keywords in the Keywords report or if your “not provided” count is still providing valuable data.
There are lots of secondary dimensions you could use to break down who your “not provided” searchers are: are they new visitors? How high is their bounce rate? How well are they converting? All this won’t tell you what search terms they’re using, but it will help you go from “I have no idea what these visitors/keywords are” to “This looks like it might be my non-brand, possibly long-tail traffic” – that is, traffic that results from many different search terms each used by just a few people. Breaking down the big, scary block of “not provided” traffic in this way will help you start to get to know the different groups of visitors you’re getting and understand something about how they behave.
To make this approach faster, create an advanced Segment for your “not provided” traffic – that is, with the condition Keyword exactly matches “(not provided)” – and then apply it to your standard Landing Pages report, or other reports, and dig deeper into how this traffic performs.
Use filters to make “Not Provided” more meaningful
If this approach proves fruitful for you, you can take it to the next level and combine the organic keywords and landing pages into a single field by creating filters for a new View in Google Analytics. This will make it easier to try out further combinations of primary and secondary dimensions to interrogate your data further. Remember, though, that filtering the entire View will pre-filter the data before any reports receive it, so make sure you create a new View and set up the filter within this, to preserve your current and original unfiltered Views. Note, too, that filters apply only from the time when you create them, so there won’t be any matching historical data to use as a basis for a like-for-like comparison.
Econsultancy provides a useful guide to creating an appropriate filter. What it does is look at the incoming search terms information, and where the search term is “not provided”, it looks to see which page that visitor landed on. Your Keywords report in Google Analytics is changed to show either the search term if it’s known or the landing page if it’s not, all within the same table. The big advantage is that you can now apply combinations of primary and secondary dimensions to address questions such as what are my likely brand visitors (homepage landers) doing compared to non-brand (specific topic page), how much traffic is buyers (product page landers) compared to researchers, and what type of content is each type of visitor looking for?
Use Multi-Channel Funnels to prove value
The last-click attribution model (see the Attribution section in the accompanying ‘Key features of your Google Analytics account’ playbook) is especially unfair to organic search, which gets no credit for bringing a customer to your site in the first place. You get a much better indication of the value of search in the Assisted Conversions report, which should include most searches that played any part in conversions at whatever stage. This report can also help you assess what proportion of your search traffic might be informational (near the top of the sales funnel) as opposed to transactional (near the end).
Hook up with Google Webmaster Tools
Integrating Google Analytics with Webmaster Tools provides the closest thing to a direct replacement for “not provided” keyword data. Once you link the two services (as explained in the accompanying ‘Getting Started with Google Analytics’ playbook), Google Search keyword information starts rolling into the Search Engine Optimization report under Acquisition in Analytics.
From here you can begin to see impressions (that is, matches) for individual keywords, clicks and click-through rates. (Bear in mind, though, that this data is for Google Search only, not other search engines such as Bing or Yahoo, so it isn’t necessarily a rounded picture.)
Here’s a tip for greater accuracy: once again, look at the Landing Pages report rather than just Queries. The latter shows only the top 1,000 queries, which might not represent the whole range of long-tail search terms driving traffic to you (remember, it’s not just the big-number terms that count!). On the other hand, it’s unlikely that you have 1,000 landing pages, so the data in the Landing Pages report isn’t getting cut off by this sampling limitation in the way that the Queries report is. This means it’s giving you a fuller picture.
Segment Organic Search Traffic by Demographics
You can see that Google Analytics offers a huge range of data about who your visitors are in chapter 6 of the ‘Google Analytics: Account Features’ playbook, and the ‘Google Analytics: Audience Segmentation’ playbook touches on ways to use this data to build up a profile or user persona. The thing to bear in mind is that you can apply audience demographic dimensions to segment your traffic data and pull out information about specific groups of visitors.
If you focus on learning more about the searchers and not just the search, you can for example find out how an age range or a gender behaves by traffic source. This is especially useful if, for example, your business appeals more to females than males: if you sell women’s accessories and a high proportion of your male search traffic land on a product page and go on to purchase, then you can infer that your site is doing a good job of enabling them to find the gifts they’re looking for.
Use Dashboards to surface the most important metrics
We’ve mentioned that you can share Dashboards, Segments and reports between accounts in Google Analytics, and even import ready-made ones from the Google Analytics Solutions Gallery – click the Browse button to view by category. Some are provided by expert analysis professionals, who may have grappled with the same issues you face. One such professional, Dan Barker, has created a website with resources dedicated to dealing with the growing problem of “not provided” search terms. Get some help from the experts!
See Paid and Organic Search Reports in Adwords
We recommend linking your Google Analytics and AdWords accounts together in addition to integrating Analytics with Webmaster Tools. Once you’ve done this, you’ll enjoy several benefits – we’ll cover these shortly, but one useful one is the ability to view the Paid & Organic Keyword report in AdWords, which marries data from your AdWords account and Google Webmaster Tools to help you understand where you have the best keyword coverage. As its name implies, this report helps you understand how often your site appears in paid and organic search for a given keyword that you are targeting.
Separate Brand from generic Paid Search terms
Another big benefit of linking AdWords and Analytics is that Analytics can automatically segment Brand and non-brand (Generic) paid search terms into distinct channels. It bases its decisions on factors such as click-through rate, text string and domain name, and you can fine-tune by switching to the Admin tab, selecting Channel Settings and clicking Manage Brand Terms. Here you can manage the list of keywords that will be included in the Brand Paid Search channel, review the terms that Google identifies as Brand and accept or decline each of them, as well as add other Brand terms that aren’t already included, such as common misspellings. Brand and non-brand paid search terms typically perform very differently, and this feature makes it much easier to analyses the two separately. Note, too, that these channels apply to all paid search, so Bing Ads and any other traffic source tagged as “cpc” will be included.Read previous article Read next article
5.4 Choosing Keywords
How to measure keywords in Google Analytics
None of what you’ve read so far should be taken as implying that keywords are a bad or superfluous thing; SEO remains a key component of a broader SEM strategy, and you’ll still want to find and incorporate appropriate keywords that will drive traffic to your site and lead to conversions.
With this in mind, let’s now look at some alternatives to keyword data analysis to help you choose keywords in a “not provided” world. (There are plenty of third-party utilities and services designed to help you find and evaluate keywords, but our focus here is Google Analytics and its related offerings.)
Put yourself in your visitor’s shoes
To develop an initial list of keywords, simply work from the perspective of someone coming to your site from the outside. The key is to find accurate, relevant answers to such questions as the following:
- What products or services do you offer? Try to focus on specific, long-tail keywords over broad keywords: if your company sells shoes, you should create a list of keywords that includes all the different types of shoes you sell. And don’t forget location-based keywords.
- What problems do your leads have that your company can help solve? If you sell waterproof iPhone cases, then visitors might be searching using phrases like “How do I waterproof my iPhone?” Use a mix of non-brand terms as well as your brand.
- How would you describe your business to someone who has never heard of your company? New users won’t now all the industry terms for your products or services, but will be searching using everyday terms they’re familiar with.
- What common questions do your leads ask? Blogs and social media can be a great way to find out.
Google Webmaster Tools
Linking Analytics with Webmaster Tools will reveal which search queries your website shows up for, with click-through rates (CTR). In a “not provided” world, the CTR is your best source of SEO performance data at the keyword level. If you also link Webmaster Tools with AdWords, you even get combined paid and organic click data for the same keyword within AdWords.
A few caveats, though: first, the figures in Webmaster Tools are calculated differently from those in Analytics, so you can’t compare them directly with historic data from Analytics, if you have any; and second, even if you sort by Clicks, the order in which terms appear is not necessarily a true indication of their importance. Treat the numbers you see as “soft and directional” rather than precise metrics.
Bear in mind that Webmaster Tools stores your data for only 90 days; if you want data over a longer period, you can download it as a CSV file and then run your own analysis in a spreadsheet.
Even if organic search terms are “not provided”, you still have access to keyword level analysis for paid search. Link AdWords to Analytics, then navigate in Analytics to Acquisition > AdWords > AdWords Keywords. The resulting table will show you your top-performing keywords, and how they relate to your conversion goals. You can make tweaks to optimize for cost per click, per lead, per conversion and other metrics, but in general ad types like Shopping Ads are very attractive to searchers who are ready to buy. Long-tailed keywords with high commercial intent are ideal for paid search, because you want your ad to be at the top of the search results and any click-throughs you receive are very likely to result in conversions.
Matched Search Query
There’s another very useful dimension in the AdWords paid search reports that you can use for SEO purposes. When you submit your keywords and bids, the search engine will match them against user search queries. In Google Analytics, you have Keywords in your AdWords report, as above, but if you create a custom report you can drill down from Keyword to Matched Search Query. The latter reveals what people actually typed – with all the long-tail variations, which is very useful for SEO.
Google Keyword Planner
The free Keyword Planner tool within AdWords (available once you’ve logged into your AdWords account) is designed to help you identify keywords appropriate for your site. The simplest way to start is to look for recommendations for a specific keyword. Click “Search for new keyword and ad group ideas” and, in the landing page part, type in the URL you’re interested in. Click the “Get Ideas” button and you’ll receive recommendations for Ad Group or Keyword. You can take these just as a list to compare with the keywords you’ve already devised by other means, or evaluate how productive they might be by looking at the Average Monthly Searches figure provided. Of course, you don’t have to restrict your analysis to landing pages only: you can use the Product Category to get ideas and data relevant to a whole area of your business.
An important caveat, though: this will focus on high-volume, fat head keywords. So, if you rely on Google Keyword Planner for your keyword research, you risk overlooking the vast range of long-tail keywords which could bring in a far greater quantity of traffic in sum, and better-converting traffic at that. Remember, too, that fat head keywords are extremely competitive and difficult to rank for.
To offset this, you could check one more factor. As above, click “Search for new keyword and ad group ideas”, enter a keyword (or list of keywords) into the field and click “Get Ideas”. Now click the “Keyword ideas” tab, and take a close look at the AdWords Suggested Bid. This is a good real-world measure of commercial intent – that is, an indication of how likely the term is to convert, in the market’s judgement. If advertisers are willing to pay a high amount per click, then you can infer that that keyword must be really valuable.
To complement this, look at the AdWords Competition column. This simply shows how many advertisers bid on that specific keyword in AdWords. The scale isn’t very detailed (Low, Medium or High), but it’s a reasonable assumption that the more people who bid on a keyword, the more lucrative that keyword is perceived to be.
Google Search Suggestions
Google Search itself can suggest potential long-tail keywords for you. Simply browse to Google Search and start to type in a keyword or phrase – a key head term, perhaps, or the beginning of a long-tail term you’re confident your visitors have used. Google’s built-in Autocomplete algorithm will offer a range of ways to complete the phrase. These appear in a pop-up; there’s also a list of Related Searches at the foot of the search results page when you’ve done a search.
The key thing to note is that these are derived from Google’s tracking of the search activity of users and the content of web pages. This means they represent the most popular long-tail terms that people have searched for recently. Some of them may be a bit off-the-wall or irrelevant to your business, and rarely-searched-for or “newly popular” terms might not generate any suggestions, but at best you’re likely to see a snapshot of what’s most searched for in your industry and you might even gain a few useful long-tail terms you previously hadn’t considered.
In theory, typing an asterisk (the Google Search wildcard character) should ensure that the algorithm looks for variations, not just the exact phrase you’ve entered (so that searching for an * a day returns results for “an apple a day”, “an avocado a day” and “an aspirin a day”), but in practice it won’t always make any difference, at least as far as Autocomplete suggestions are concerned. Note also that unusual spellings are likely to get normalized, so you’ll see suggestions (and search results) relevant to the most common spelling of the word Google thinks you’re after. Finally, be aware that it used to be possible to type a tilde (~) character before a term to search for similar terms and synonyms as well as the exact word (as in ~footwear) but Google appears to have discontinued support for this.
If you’d like more than the handful of suggestions Google’s Autocomplete algorithm offers, try visiting ubersuggest.org, which trawls a range of online suggest services to generate a long list of potentially related terms. You could even search thesaurus.com for words like your starting term.
Google has another trick up its sleeve: Google Trends can help you discover what search terms are most popular and fastest-rising. By default, it shows you the most-searched terms now, in a specific country or in a specific year. Top Charts present lists of people, places and things ranked in order of search volume, while Hot Searches highlights queries that have jumped significantly in traffic based on real-time monitoring. More usefully for keyword research, you can find Top Searches related to the category, country or territory you’ve chosen, and even enter a term to find Top Searches similar to it.
You can specify an exact phrase by entering it in inverted commas, and even include variations or misspellings of a term, if you wish, using a plus sign – so for example tennis shoes will find searches including both tennis and shoes in any order, “tennis shoes” will find searches for the exact phrase, and tennis + tenis + tenniss will give you data on searches including any of these spellings. You can restrict your search to specific countries, periods and categories using the drop-down menus under the search box.
To help you evaluate potential keywords, the Top Searches will show you how many searches there have been for similar terms. Rising Searches is potentially more useful; it shows related terms that have grown significantly in popularity over a given period when compared to a preceding period. For each rising search term, you see its percentage growth over the period (or just “Breakout” if the growth is greater than 5,000%). Note, though, that only the 10 most popular top or related searches are displayed – to see more, you’ll need to click the Google gear icon and download a more complete data set as a CSV file, then open it in a spreadsheet.
Where Google Webmaster Tools focuses you on clicks and Keyword Planner helps you with keywords to target by landing pages, Google Trends is valuable because it helps expand your keyword portfolio (top searches) and shows you the keywords in your sector under which you should be considering lighting a fire (rising searches).
It’s even possible to subscribe to Google Trends notifications, so that it automatically sends you an email if there’s a significant increase in search volume for any search topic, Hot Searches for any country, or any US monthly Top Chart. You can subscribe to notifications about trending topics by location, so for example you can keep on top of hot search keywords in the UK even if you live in New Zealand, or vice-versa – you simply choose the topic, the country and how often you want to receive notifications. Alternatively, you can receive a regular digest of all the top searches within a specified country.
It may be that a lot of visitors arrive at your site by looking for your brand or another broad search term and then narrow down by searching within your site. If so, could you be attracting more users more directly, or making life easier for your present users, by ranking higher for the ultimate destination search terms? If you offer a site search function (and you should!), then you can mine it for details of the long-tail search terms your visitors are using to find what they’re ultimately after. As Google puts it, “each time users search your site, they tell you in their own words what they’re looking for”.
You first need to follow Google’s guide to setting up site search tracking – it can be quite technical, depending on how search is implemented on your site. Once it’s in place, you can find your Site Search reports in the Behavior section of the Reporting tab (see section 3.3 of the ‘Google Analytics: Key Account Features’ playbook). The Site Search Overview report includes a list of the top search terms used, or you can drill down in the Search Terms report to see the keywords entered into your website’s search box, with metrics for the total number of searches, % search exits and additional details about visits related to a search term, including the number of times users viewed a search results page after searching (Results Pageviews/Search) – a count higher than 1 or 2 can indicate that visitors are having to work hard to find relevant results (it also indicates their determination to find that item!). So, look closely not just at the actual search terms for potential keywords, but evaluate how well your site delivers on each of them.Read previous article Read next article
5.5 Benchmarking Competition
How to audit and benchmark against competitors
There are quite a few tools out there for competitive analysis of search performance, although some experts warn that these too will become less reliable as the proportion of “not provided” search terms grows.
Within Google Keyword Planner, if you’re analyzing your own website, you can see your Ad Impression Share. Knowing how often (relatively) your ad shows up for each keyword can help you refine your keyword strategy, although it’s inevitable in highly competitive markets that there are likely to be fluctuations as your competitors try to do the same thing.
Search Position Finder is one of a number of free tools that can be handy when you’re considering a keyword and you want to find out where your website or a competitor’s ranks on Google Search (or other search engines) when that keyword is used. You can however search for only one term at a time and see the result for only one website at a time, so it doesn’t deliver very much more than simply doing a Google Search for that term yourself.
SERP Checker is more explicitly designed for competitive analysis. You enter your domain and up to three competitors, and the site quickly checks the top 300 search results for your site ranking for up to 25 keywords at a time. If you register, you can add up to 20 domains with 25 keywords each, making a total of 500 keywords, and graph results for the last 30 days. For further analysis, you can export results to investigate in your own spreadsheet.
KeywordSpy is a subscription-based service offering competitive keyword tracking in AdWords and other PPC campaigns, but it is possible to simply enter a domain or a keyword on the website and view a range of competitive information, including competitors and their rankings (for both paid and organic search) and top keywords (paid and organic) complete with volume and CPC data. Sign up for a “lifetime free trial account” and you can access more than the top 10 results in each category.
Enter a URL on the SEMrush website and you can view the site’s top five organic and top five paid keywords plus some competitor information, but for detailed competitive analysis you need to subscribe. SEMrush also provides organic keyword referral data, updated monthly, returning those organic keywords for a webpage which are in its database and for which your webpage ranks in Google’s top 20. SEMrush says it analyses the rankings of the most profitable and popular keywords – over 95 million keywords – but even a number this huge can’t include all the long-tail terms that might be driving traffic to your site.
SimilarWeb is also a paid service, but you can enter a URL on the site and view a wide range of information about it including organic and paid keywords, traffic overview and engagement figures, and much more, all presented in very accessible visual form, and download all this as a PDF. In the free online version, you can enter one competitor site to compare much of this data with, though not specific keyword performance.
5.6 Linking Google Accounts
How to link Google Analytics to Google Adwords
Linking Google Analytics and AdWords brings benefits to both tools, but it’s specifically useful for SEM because it gives you a more detailed picture of the effectiveness of your paid search strategy than AdWords alone can do. You can for example view Google Analytics site engagement data in AdWords, and automatically view your AdWords click and cost data alongside this, so you can evaluate cost and return on the basis of more than just sales performance. To the same end, you can import Google Analytics Goals and transactions into AdWords as conversions, and create remarketing lists in Analytics to use in AdWords for targeting specific audiences.
How to link Analytics to Adwords
The linking wizard makes it easy to link your AdWords account(s) to multiple Views of your Analytics property. If you have multiple Analytics properties and want to link each of them to your AdWords account(s), just complete the linking wizard for each property. You can start the process from AdWords or Analytics. We’re focusing on the latter here, so we’ll start within Analytics. First make sure you’ve signed up for both Analytics and AdWords using the same Google Account, and that this account has Edit permission for the Analytics property and Administrative access for the AdWords account(s). Then just follow this guide:
How to link your Analytics account to Adwords
1. Start linking process
Click the Admin tab, select the relevant Account if you have more than one, then click the Property you want to link. Click AdWords Linking.
2. Select accounts to link
Tick the box next to any AdWords accounts that you want to link with your Analytics property. If you have a My Client Centre (MCC) account, expand the MCC account by clicking the arrow next to it, and then tick the box next to each of the managed AdWords accounts that you want to link.
3. Link configuration
Click Continue, then, in the “Link configuration” section, enter a title to identify your group of linked AdWords accounts. (You won’t need more than one link group unless you have lots of AdWords accounts and want data to flow in different ways between these accounts and your Analytics property, for example if you want to enable auto-tagging for only some of them.)
4. Choose Views
Select the Analytics Views in which you want the AdWords data to be available (see section 7.6). Bear in mind that anyone with access to that View will be able to see your imported AdWords data and, in the same way, if you choose to import Analytics data (such as Goals and transactions, metrics or remarketing lists) into your AdWords account, anyone with access to that AdWords account will be able to see your imported Analytics data.
5. Manual tagging options
The account linking process will enable auto-tagging for all your linked AdWords accounts. Click Advanced Settings only if you need to manually tag your AdWords links (you almost certainly don’t).
6. Complete linking process
Click the Link Accounts button, and you’re done. With auto-tagging turned on (as is recommended), Analytics will start automatically associating your AdWords data with customer clicks.
AdWords is great at telling you how much money your ads cost and how many sales you’re getting, but what happens between the click and the conversion remains a mystery unless you add Google Analytics to the equation. It can tell you what people do on your site, so if they’re not converting, it can help you find out why. Look at the Bounce Rate and Pages per Visit figures to see if visitors aren’t finding what your ads were promising. Use the Goal Flow report to find where your AdWords clicks are dropping out of the conversion process. Even find the best performing position for your ads with the Keyword Positions report.
If you’ve assigned a value to a Goal (which may be a lead rather than an actual purchase, for example) then Google Analytics will display Revenue per Click (RPC), Margin, and Return on Investment (ROI). Click Acquisition in the sidebar and select AdWords > Campaigns to view these. Above the graph, select Clicks to see your raw AdWords data.
One of the most interesting comparisons is Revenue per Click for your AdWords compared to Revenue per Click for the site on average. This number is located at the top of the column right by the header. This is not an infallible measure of the “value-add” of your AdWords campaign, but it’s a strong indicator: if it’s not bringing in substantially more revenue, is your spend worthwhile? (Or conversely, are you sending pay-per-click visitors to the appropriate pages, or is the site failing to deliver on what the ads promise?) The ROI column is more accurately “Return on Ad Spend”, and if sorted from low to high, can quickly point you towards your AdWords campaigns that need the most immediate attention. Again, the changes might need to be made in your ad text or content, or it could be something more to do with the landing page associated with these ads.
6.1 Analytics for CRO: An overview
Analytics for website and conversion rate optimisation
We’ve already noted that attracting visitors to your website is only the beginning. Mere footfall is rarely – if ever – your sole business goal. Measurement comes in terms of macro conversions – in other words, end goals such as a user making a purchase on an e-commerce site or submitting a contact form on a lead generation site – and micro conversions (steps along the way, such as adding an item to the site shopping basket).
As well as measuring conversions, Google Analytics includes a wide range of ways to investigate what visitors are doing on the site, which can give you invaluable clues to where you might need to focus your attention and how you can smooth the road towards conversions. This is what site optimization is all about. It focuses on the user experience, but not for its own sake. The point is not simply to make your site visually attractive or fun to use, though there may be nothing wrong with doing so. The aim is to ensure that the site serves your business objectives as well as possible and that the business will benefit from users reaching their goals on the site.
Site optimization for CRO
One of the best ways to ensure that your site is optimized for conversions is to apply a user-centric approach to design and segmentation, which is formally referred to as user experience design or UX design. It’s actually about applying the same principles as an intelligent marketing strategy. In your marketing activities, you create and share information that you know is of interest and value to your target user persona. You keep this persona’s interests, preferences and needs at the fore of your planning, creation and distribution processes while, as we’ve seen, keeping what users are looking for at the heart of your SEO activities. In the same way, UX design helps you achieve your website conversion goals by creating a design that is focused on the user’s objectives and removes any barriers that might inhibit their journey.
The elements of this design include visual design and accessibility, naturally, but also “information architecture” (that is, the site structure and navigation), usability (more about this in a moment) and user interaction and journey analysis. With well-defined, efficient user journeys mapped out, you can concentrate on delivering the right information for that user at the right time – and strategically implement conversion points at the most appropriate moments along the way. It’s evident how the Users Flow and Behavior Flow reports in Google Analytics can be useful for discerning and defining the paths your users take through the site and the conversion process, as well as identifying problem spots where they might be dropping out.
Website usability is not about attractive design, visual impact or “coolness”, it’s all about effectiveness – that is, how effective the site is at enabling specific users to complete specific tasks. If they fail to do so, where do they hit the critical stumbling block? Site optimization involves identifying these roadblocks (pages with high exit rates) and determining how they’re falling down on the job. If users do complete the desired task, how quickly and easily do they do so? The various Flow reports in Analytics can help you determine how many steps they took, how much of their time they spent on hunting through the site or seeking help (FAQs, site map or site index, site search) and how much was productive time. If users resort to site search, don’t assume it’s necessarily a failure of site design and navigation, though: some users prefer searching and find it less time-consuming than browsing for what they want.
Bear in mind that users accessing your site on mobile devices have different needs and may have different objectives. Typically, they want information in quick, easily digestible bites, and more than half will abandon a site if it takes more than three seconds to load. Some 30% will abandon a purchase transaction if the shopping cart isn’t optimized for mobile devices. It’s worth catering for these fussy visitors, though: these days a quarter of web searches are conducted on a mobile device, and 80% of shoppers admit that mobile purchases are impulse-driven.  They add that they’re more likely to purchase from a brand that offers an engaging mobile experience, though the eventual conversion might be at a later point on a desktop computer. Using a responsive design or creating a version of your site for mobile devices will help ensure you attract and keep these users; by making the path to purchase or other conversion simple and straightforward, you’ll align the site better with their needs and likely increase its conversion rate.
The site speed will affect a visitor’s experience of your site: the longer a page takes to load, the higher its bounce rate is likely to be. A whole range of speed measures is available in the Site Speed reports in Analytics (see section 3.3 of the ‘Google Analytics: Key Account Features’ playbook), but it’s worth noting that Google Search itself takes page speed into account – along with other factors – when deciding search result rankings. If you hope to improve the site’s ranking, you need to look carefully at three specific factors: response time (time to first byte), page size and page load time.
From the user experience perspective, a page that loads (or an operation that completes) in one tenth of a second (100ms) or less will feel instantaneous to the user, so this is the “gold standard” to aim for. In these impatient times, when we’re used to super-fast broadband speeds, five seconds is probably the upper limit. Sometimes, however, it’s just not technically possible to complete an operation in 100ms – sorting a large data table, for example. In these cases, experts recommend using some sort of loading dialog to show the user that something is happening and the site hasn’t just locked up. Animations such as spinners are good, but progress bars are better, and a percentage readout display is better still.
It’s useful to optimize page assets and load time, but when it comes down to it, perceived performance can be more important than actual performance. For the same reason, it’s vital to ensure that clickable items on a page respond to mouseovers to signal that they are indeed clickable, and change state when they are clicked to confirm the user’s action. Give users as much feedback as possible and they’ll feel that your site is more engaging.
Indeed, it’s well worth looking closely at how visitors interact with your website. The In-Page Analytics report under Behavior in Analytics can help establish exactly how users are interacting with specific pages. Where are they clicking on the page? What percentage of visitors is interacting below the page fold? If you find that users are drawn to click on specific areas of the page, consider moving your most important links – that is, calls to action or other links that you know are more likely to lead to conversions – to that area, even if this means rearranging navigation menus or page structure. If you find that users are tending not to travel below the fold, try moving any conversion-critical elements to a more prominent position on the page.
In the Site Search reports under Behavior, you can find out what visitors are searching for using your on-site search (Search Terms) and where they’re searching for it (Pages). Are there patterns in what they’re looking for? Consider adding that content to the site, or creating clearer navigation to it if it already exists. Do they often search for the same thing from a specific page? This can alert you to possible shortcomings on that page. Give users what they’re after, or make it easier to find, and they’re that much less likely to want to go elsewhere for it.
6.2 What is CRO?
What is conversion rate optimization?
We’ve noted that – depending on the goals of your site – a conversion might be a purchase, but could also be the creation of an account, the completion of a survey or contact form, signup for an email newsletter, an app download, and so on. No matter what the objective is, though, you want as many of the site’s visitors as possible to achieve that objective (or even several – perhaps a succession of micro conversions leading to a macro conversion). Conversion rate optimization (often abbreviated to “CRO”) simply means taking a structured and systematic approach to increasing the proportion of site visitors who achieve a conversion.
Measuring the engagement of your visitors is an important first step. A high bounce rate and low Average Time on Site means visitors are probably not sticking around long enough to do whatever it is you want them to do. Average Page Views is also an important engagement metric, but you need to relate it to your site’s specific goals – if visitors are viewing a high number of pages but eventually leaving without converting, this can mean a lack of clarity in your conversion funnel.
A key technique in CRO is A/B testing – that is, trying out variations of pages and measuring which produce a higher conversion rate. Just about any element on a page can be tested, from the color of a call-to-action button to the actual product copy, image size, layout, amount of text, fonts used – anything. And as we’ve seen, Google Analytics includes built-in Experiments reports under Behavior, which make it possible to set up as many as 10 variants of a page and track their results against a whole range of metrics. There’s a bit of setting-up involved, specifically in creating variant pages each with a distinct URL, but when you click “Create experiment” in Analytics the wizard makes it relatively straightforward to configure the testing and specify which metrics you want to apply.
Bear in mind that Google Analytics offers a wide range of ways to measure conversions and investigate how users achieve them. In the Conversions reports you can define and track Goals and Ecommerce performance, and you can also view Events reports under Real-Time and Behavior, enabling you to take a close look at what users are doing. You can trace their paths through your conversion funnel using Flow visualizations and Multi-Channel Funnels.
As one expert puts it, remember that the goal of CRO is not to manipulate visitors into converting.  It’s to ease the journey of already interested or engaged visitors through your website until they’ve achieved the outcome they desired themselves. If a user has searched for “blue Nikes” and landed on your product page, chances are they want to purchase the product. It’s not trickery to make doing so as simple or even enjoyable as possible. Removing barriers, simplifying forms, clarifying navigation, all these things lead to an improved customer journey and a better user experience. That customer is more likely to come back for future purchases and recommend you to other users.
CRO makes sense because it makes the most of the traffic you’re already getting. You aren’t spending more money getting visitors to your site, just doing a better job of converting them once they get there. Optimization increases the return on your current investments in SEO or paid search, and converting a higher percentage of your current visitors is likely to be much more cost-effective than attracting new ones. It reduces your customer acquisition cost and improves the bottom line.
6.3 Goals for CRO
How to set goals for CRO via Analytics
We’ve mentioned that website and conversion rate optimization involves ensuring that the objectives for which users are visiting the site align with your business objectives for the site. We’ve noted, however, that there may be different types of objectives, and “conversions” are not necessarily purchases. So, let’s briefly recap the types of Goals that you can define and track in Google Analytics. As we saw when we looked at the Conversions reports, these are additional to purchases, which you track by enabling Ecommerce reporting and adding the required code to your site or Google Tag Manager.
What goals can be set to measure CRO?
In Google Analytics, you can define four types of Goals for a site.
- Duration: a session lasts a specified time or longer.
- Pages/Screens per session: a user views a specified number of pages or screens.
- Destination: a user arrives at a specified location in the site – that is, a specified page or screen loads, such as a sign-up thank-you page.
- Event: a specific action occurs, such as a button click or the play of a video.
To set up a Goal, click Admin, then select a specific Account if you have more than one. From here choose the Property and select a View. Now click Goals, then “Create a Goal”. Simply follow the steps before clicking “Save Goal” to finish.
You can use any of the common use case templates as a starting point or create a custom definition. Note, though, that the templates are industry-specific and you won’t see any if you haven’t specified an Industry Category for the Property. Alternatively, instead of clicking “Create a Goal”, you can import Goals from the Google Solutions Gallery and use them as-is or (more likely) customize them to suit your specific requirements.
Before clicking “Save Goal”, you might wish to click Verify to test your Goal setup. All this does, however, is test whether there’s data available in that View that matches the criteria you’ve defined for the Goal. It does not confirm that the Goal setup will perform in the way you expect or deliver the data you want.
Goal data starts being collected as soon as you save a Goal, and you can test things yourself by visiting your site and performing the actions defined in the Goal. Then, in Analytics, navigate to the Conversions reports under Real Time to see whether your “conversion” has been recorded.
There are a few things to note about Goals that you don’t need to go on a fancy seminar to understand. The most important consideration is those Goal conversions are counted only once per session. This is logical enough in the case of Duration or Destination Goals, but bear in mind that it applies equally to Event Goals, so that for example if a visitor plays the same video five times in the same session this is still counted as just one conversion for the video play Event Goal. The same visitor can, however, complete more than one Goal within a session: for example, if Goal #1 is “watch a video” and Goal #2 is “complete the sign-up form” and the visitor does both within the same session, then this will count as a conversion for both Goal #1 and Goal #2.
It’s not possible to mix different types of criteria within one Goal – a Destination Goal, for example, concerns only the destination and not how long the user took to get there, so the Goal can’t measure whether the destination was reached within a given time or number of steps. You can, however, use dimensions such as Time on Site and Page Views as secondary metrics in various reports, and use the Goal Flow report to analyses the exact path that your visitors travelled through on their way towards a Goal conversion. Once you’ve set up Goals, the number of conversions coupled with conversion rate (plus Goal values, if you’ve assigned any) will appear in the Conversions reports, and Goals also appear as a sub-tab in all the Analytics reports that will allow you to focus in on your Goal data.
Finally, note that it’s possible to switch Goal Reporting off for the entire View, in which case Goal related data is not collected and Goals are hidden in all the reports in that View. Goals can also be switched off individually, in which case no data will be recorded for that Goal. You can’t delete Goals once they’ve been created, but you can redefine them by changing their criteria, in which case the relevant data will be collected from that point onwards; any historical data collected under the Goal’s previous definition remains unchanged. In all these cases, beware of glitches or gaps in your data!
6.4 Analytics and conversion tracking
How to use Google Analytics and Adwords Conversion Tracking
If you use Google AdWords, you may be aware that AdWords itself can track conversions that occur after the user clicks your ad. Here too a “conversion” doesn’t have to be a purchase; it could be a visitor signing up for an email newsletter or filling in an enquiry form for more information – whatever you decide is significant for your site and your business. It is, however, less sophisticated than using Analytics. When users click on your AdWords ad on google.com or selected Google Network sites, a temporary cookie is placed on their computer. You add a snippet of code to the thank-you or confirmation page that users see after they complete your desired conversion, whatever it may be. This confirmation page code interacts with the cookie, and the conversion is recorded. If you want to track multiple conversions, you’ll need a confirmation page and a snippet of code for each one.
You can choose between two counting methods for each conversion action you’re tracking. “All conversions” counts all the conversions that happen after an ad click, so that if, for example, a user arrives via an AdWords ad and buys three items on your site in separate transactions, this is counted as three conversions. “Unique conversions” counts each type of conversion only once, so that if a user fills in two separate enquiry forms for information and signs up for a newsletter, this is counted as one enquiry conversion and one sign-up conversion. You can choose to use a different counting method depending on the type of conversion action – typically you might choose “all conversions” for sales, where you want to track the actual number of sales you get from AdWords-led visits, and “unique conversions” for enquiries, where you’re not interested in the total number of enquiries as much as in the number of people who make enquiries, or you want to know not the total number of leads but whether or not a certain kind of lead has resulted.
How to set up conversion tracking in Google Analytics
To use conversion tracking in AdWords, you first need to generate the required code snippet for each confirmation page.
1. Create conversion
Sign in to your AdWords account. Switch to the Tools tab, and select Conversions from the drop-down menu. In the “Conversion actions” tab, click the “+Conversion” button.
2. Supply name and source
From the Source menu, select where the conversion will occur – here we’ll assume it’s “Webpage”. In the “Conversion name” field, enter a name for the conversion you want to track – for example “newsletter sign-up” or “footwear purchase”.
3. Choose category
Select the most appropriate Conversion Category – Purchase/Sale, sign-up, Lead, View of a key page (for example your Contacts page) or Other. Note, though, that what you select here has no impact on how conversions are recorded.
4. Track conversion time
From the Conversion Window menu, select how long after an ad click you want to track conversions for this conversion action. You can choose anything from a week to 90 days – longer periods are ideal if your conversion funnel involves several steps and users tend to visit the site several times before converting. The default is 30 days.
5. Select what conversions to track
From the Count menu, choose whether to count all conversions or unique conversions for this conversion action, as explained above.
6. Set conversion value
In the Conversion Value field, choose whether each of these conversions has the same value, varying values or you’d prefer not to count a value. If you’re selling different products at different prices, you can set up a tracker to record the prices in your shopping cart, but this is an advanced option that requires some expertise and varies according to different e-commerce platforms. When it’s set up correctly, though, any conversion value statistic, such as total conversion value and cost per click, will reflect the actual revenue from the products you’ve sold instead of a fixed value.
7. Final touches
To finish, decide whether you want to put a Google Sites Stats notification on the page to notify visitors that you’re using conversion tracking. Skip the advanced options for now, and click “Save and continue”.
You’ll now be asked whether you make changes to your site code yourself. If you answer yes, the required code will appear in a new window. Copy it and paste it into the code of the appropriate confirmation page on your site. If someone else controls the site code, enter their e-mail address and the code snippet will be sent to them.
Once tracking is up and running, you can see information about your conversions from your AdWords account’s Campaigns tab at the Ad Group, Ads and Keywords levels. A Conversions column displays the count for each conversion you’re tracking (counted using whichever method you’ve chosen for each one, “all conversions” or “unique conversions”), and a Converted Clicks column displays the number of AdWords ads that led to any conversions (regardless of the value of those conversions – there’s no distinction between high-value conversions such as multiple purchases and low-value conversions such as a single newsletter sign-up). Other columns include Conversion Rate (the percentage of ad clicks that resulted in conversions) and Cost per Conversion (how much you spent on clicks divided by total conversions).
Advanced options include iOS app tracking (but only for ads served in mobile apps through the Display Network) and View-Through Conversions, which tracks when a person sees your ad but doesn’t click it, yet visits your site later and completes a conversion – in other words, they might have been influenced by your ad to return and convert later. You can set this to track for a specific period in between the view and the conversion, from 1 to 30 days, or a custom amount.
Even sticking to the basic reporting, however, you can determine which ads, keywords and campaigns bring in business, and compare how much. For example, say you find that visitors who click on your ad with “buy designer jeans” as a keyword buy a lot of jeans. Meanwhile, you see that a few people click on the ad with “blue jeans” as a keyword but none of them make a purchase. Both ads are bringing traffic to your site but the latter is leading to conversions, so you know which is delivering a better return on your investment.
Once you’ve been tracking conversions for some time and enough data has been collected, you can use Search Funnels to find more detail such as how much time elapsed between the first time a user clicked on your ad and when they completed the conversion, and how many times they saw your ad before converting.
How to import Analytics goals and transactions
Linking makes it possible to view Analytics site engagement metrics alongside your AdWords performance data, giving you a more detailed picture of what visitors are doing on your site.
The metrics available include Bounce Rate, Average Session Duration and Pages per Session, which can indicate whether visitors coming from AdWords campaigns are finding what they expected on the site. The Percentage of New Sessions metric can help you assess how effective your campaigns and ad groups are at attracting new visitors.
Even more usefully for our present purposes, you can import your Google Analytics Goals and transaction information into AdWords Conversion Tracking. This is useful because Analytics can handle different types of conversions, such as Events (playing an embedded video, for example) that don’t result in the appearance of a confirmation page and hence can’t be tracked in AdWords.
To begin importing Analytics Goals, first make sure your Analytics and AdWords accounts are linked. You’ll also need to have enabled data sharing in your Analytics account (see section 4.1 of the ‘Getting Started with Google Analytics’ playbook) and auto-tagging in your AdWords account. Finally, each type of Analytics Goal or transaction won’t appear in AdWords reports unless it has received active traffic from an AdWords ad at some point.
How to import your goals
1. In your AdWords account, open the Conversions page.
2. In the Conversion Tracking table, click “Import from Google Analytics”.
3. From the list, select the Goals or transactions that you want to import, then click Import at the bottom of the table.
AdWords Conversion Tracking starts importing the data from your Analytics account from that day onwards, and won’t import any earlier data. Note that it may take up to two days for AdWords to import your Analytics data.
When you import Analytics Goals, you can choose whether they’re counted in AdWords on an “all conversions” or “unique conversions” basis. If you import the same Goals into different AdWords accounts, you can choose a different counting method for each account.
Your imported Analytics Goals appear alongside the conversion data in your Conversions page and AdWords reports within two days. Imported Goals have the same names as in Analytics, with the name of the View in parentheses after this. For example, an Analytics Goal named “Sign Ups” located in a view called “Master View” is named “Sign Ups (Master View)” in AdWords Conversion Tracking. The name can’t be changed in AdWords, only in Analytics.
If you’re not seeing all the extra Analytics data you expected in AdWords, you might need to add new data columns to your AdWords reports. In your AdWords account, click the Campaigns or Ad Groups tab, as relevant, then click the Columns button and select “Customize columns” from the drop-down menu. Look for Google Analytics in the left-hand column, and click this to see the available metrics. Click Add for each column you want to add, and click the Apply button to finish.
Finally, be aware that Analytics and AdWords count some things differently, so there might be discrepancies in their data. For example, Analytics counts sessions while AdWords counts clicks, so if a user finds your site by clicking on an ad, bookmarks it and later returns to the site via the bookmark, AdWords records this as one click but Analytics counts it as two sessions. There can also be a time lag in importing data from Analytics into AdWords.Read previous article Read next article
7.1 Analytics Metrics: An Overview
How to choose your Google Analytics Metrics
Google Analytics offers dozens of reports, each containing a multitude of metrics. This enables you to measure and analyze just about any aspect of your site’s traffic and performance that you can think of… Or just get mired in too much information. How do you separate the signal from the noise and pick out what you really need to know about?
What relevant metrics can be measured in Google Analytics
The key to using Google Analytics effectively is deciding upfront what your key business objectives and desired outcomes are, and how you can measure these. Then, instead of trying to measure everything possible, create a simple framework aligned with your desired outcomes and objectives. In other words, match your metrics to your business needs and measure what matters.
So, if you run an e-commerce site, the most valuable information will be such things as which traffic sources lead to the most conversions (or the most profitable conversions) and whether there’s any step in the sales funnel at which you’re frequently losing customers. For information or help sites, you’ll want to measure engagement in terms of how long visitors spend on the site, how many pages they visit and whether they talk about the site on social networks. For brand-building and promotional sites, you’ll want to know how visitors found the site, so you can decide where best to invest further efforts – social marketing, paid search or organic search?
Measuring which pages draw the most traffic can help you assess whether you’re providing enough of what your users are looking for. And so on. It’s clear that some metrics can be applied across industries, but your final choice of metrics will depend on your goals: consider what you need your website to do for your business and then craft your reporting to deliver the information you need.
Here we’ll focus on a few of the key performance indicators likely to be relevant to a range of business types and briefly discuss how they could help you.
Users and Sessions
These are the fundamental measures of activity on your website or app: how many people use it, and how often they do so. In web-only terms, you’d be talking about visitors and visits, and those are the terms Google Analytics used until April 2014, when it adopted the terms users and sessions respectively in order to cover apps as well as sites. You can find out a great deal about your users in the various Audience reports, including their age and gender, their geographic location and many aspects of their activity on the site.
Learning who is coming to your site, where they’re located and what their interests are, among other things, can help you assess your site’s strengths and weaknesses in attracting new visitors, evaluate whether your content is appealing to those visitors, and develop strategies to convert them into customers.
New to Return Visitor Ratio
The first time a device loads your site’s content and a hit is recorded, Google Analytics creates a random, unique ID that is associated with the device. (In practical terms, a cookie is set in the visitor’s browser, which persists for two years.) This unique ID is sent to Google Analytics in each hit, and if Analytics recognizes it as an existing ID, it counts a Returning User. If it’s a new ID, it counts a New User.
In some reports, it’s possible for the same user to be double-counted. If you set a date range of 30 days, say, and a user pays a first visit to the site and then returns later, both within the 30-day period, then Analytics will record two sessions, one with a new user and one with a returning user, even though both are the same person. In the same way, for some dimensions such as Source or Medium, it’s possible that the same unique user can be in multiple buckets – for example, if a user visits from both organic search and paid search within the same date range. For this reason, when you view Users over such a dimension, the sum of the rows won’t add up to the correct total. Conversely, if a user has cleared their cookies, they’ll be counted as a new user. The same applies if they visit again using a different device, unless you’ve implemented the User ID feature, which enables you to assign your own unique IDs to users to identify them across multiple sessions and devices – see chapter 7 of the ‘Key Account Features of Google Analytics’ playbook.
With these caveats, it’s very useful to know that you’re attracting new visitors and expanding your potential customer base. As a rule, however, returning visitors are much more likely to convert – they know your brand or site and they’ve chosen to return, so you’re probably providing something they like. They’re also more likely to promote your brand on their own social networks. They might not convert on their second or even third visit, but it’s important to remember that it can take several points of contact to make a sale.
The Real-Time reports in Google Analytics, but particularly the Acquisition reports, can tell you where your visitors are coming from. Are they finding you by organic search, paid search (your cost-per-click or CPC campaigns, such as Google AdWords) or referral from other sites or blogs? Are they coming from social networks such as Twitter or Facebook, as a result of your newsletter activity, or by typing in your URL directly? Knowing how visitors found your site is a valuable guide to what’s working for you and what you need to work on. Correlate source with conversions and you can gauge whether your investment in paid search or other forms of marketing is paying off by bringing in high-value customers.
If a visitor arrives on your site and then leaves from the same page without interacting with the site, this is a “bounce”. In addition to viewing the simple bounce rate in the Behavior reports in Google Analytics, you can analyses bounce rate in connection with many other dimensions to assess whether users are finding what they want on your site – correlate with geographic location, for example, and you might conclude that users from a specific country could need more localized information or content in their language. You can see which pages have the highest bounce rates in the Site Content > All Pages report: are there trouble spots? Or, correlate bounce rate with Site Speed: are there pages that load so slowly, your visitors can’t be bothered waiting for the content?
A high bounce rate may be an indication that your content isn’t engaging, or that your advertisements are misleading and visitors aren’t finding what they expected to find. Beware, though: a high bounce rate doesn’t always mean your visitors are dissatisfied. If a website is so well designed that a customer lands on the page, finds what they want, takes action, and then leaves without visiting any other part of the site, the bounce rate would be high. For a help site, for example, short visits could indicate that your content is doing its job perfectly. So always correlate with other metrics, and if necessary build in cross-checks such as feedback forms to verify whether your visitors are leaving because they’re unhappy or because they’re satisfied.
Time on site
Average Session Duration is a measure of how long visitors (of all types) spend on your site. Like a high bounce rate, a low time on your site doesn’t necessarily mean your visitors are dissatisfied; it could just mean they’ve found what they wanted quickly. This is ideal if your site is there to provide help or advice. It’s not so good, however, if the site offers in-depth content and visitors aren’t staying long enough to consume it, still less to convert (whether that means buying something or filling in a form). Do you need more engaging content? Implement Event Tracking (‘Key Account Features of Google Analytics’ playbook, page 39) to check whether visitors are interacting with your site at all. Correlate with other measures of engagement, such as return visits and social referrals: if they like what they’ve found, they’re likely to come back again and talk about the site on social networks.
You can’t always take high Average Session Duration at face value, either: sometimes people will leave the browser window open without interacting with the site. This will drive up the time spent on the site artificially. So once again, correlate with other measures of engagement, particularly appropriate Events such as clicking to activate content or moving to the next page.
This is a key metric for any type of site and no matter how you define “conversion”. As the ‘Key Account Features of Google Analytics’ playbook (section 3.4) reveals, you can define your site’s Goals in whichever terms are appropriate to the aims of the site, whether it’s sales, sign-ups or anything in between – even number of pages viewed. Once this is set up, you can evaluate almost every other metric in relation to conversion rate. The Acquisition reports in Google Analytics (‘Key Account Features of Google Analytics’ playbook, section 3.2), for example, might tell you that organic search is bringing in many times more users than paid search, but if the users who arrive via paid search go on to view more pages and have a far higher conversion rate, then your investment in paid search is actually doing its job. Goal conversions are the primary metric for measuring how well your site fulfils its business objectives, and this measure can be applied at every level to determine whether a single campaign or an entire traffic source, a specific page or a specific product line is delivering value and contributing to your bottom line.
Revenue Per Click
Paid marketing can be a substantial investment, and Google Analytics can help you assess how worthwhile it is. The Acquisition > Cost Analysis report shows session, cost and revenue performance data for your paid advertising campaigns, including any non-Google marketing channels for which you upload cost data, plus AdWords (labelled “google/cpc”) if you’ve linked your AdWords and Google Analytics accounts (‘Google Analytics for SEO, SEM, Website and CRM’, section 2.5) and imported AdWords cost data to the View you’re using. The report compares the cost of each campaign with its associated revenue (from your actual e-commerce performance or the fixed value you assigned to specific Goals) to calculate ROAS (Return on Ad Spend) and RPC (Revenue per Click).
Comparing your RPC with the Cost per Click is a quick-and-easy measure of whether your investment is bringing in a good return. Similarly, it’s quite straightforward to compare the ROAS from AdWords with that from other channels or simply with the average for the site, although this isn’t necessarily the only way of judging whether AdWords is sending the right leads to the site: if it (or any other channel) seems to be underperforming, you might need to look also at the wider picture – your campaigns might not be sending pay-per-click visitors to appropriate pages, or your ad text or site contents might need some attention.
Cost Per Conversion
Calculating the return on your ad spend (revenue per click minus cost per click) gives you a basic measure of the value of your paid marketing, but this is not likely to be the whole story. Cost per conversion has been dubbed “the only metric that matters” because you can find your actual profit and true return on investment only by comparing your actual revenues per conversion with the real cost of conversions. Neither of these, however, is as straightforward as the click-based measures we’ve just mentioned. Your revenues per conversion might be actual e-commerce revenues or nominal values you assign manually to specific types of conversions, in which case you need to give some consideration to ensuring those assigned values are realistic; and the cost of conversions ought to include your whole range of costs – not just the amount you spent on paid advertising, but the costs of implementing other forms of marketing, content creation and so on.
This is important because typically many factors contribute to a conversion. The path that a customer takes to reach a conversion might include several touchpoints, each of which contributes in some degree, so you’ll want to begin by looking at the Multi-Channel Funnel reports in Google Analytics (‘Key Account Features of Google Analytics’, page 46). In particular, the Assisted Conversions report shows how many sales and conversions each channel initiated, assisted and completed, along with the value of those conversions and sales. You then need to decide how much credit to give to each channel or type of touch point, and apply the appropriate attribution model to reflect the contribution of each. It’s only after you know this that you can calculate the corresponding cost distribution.
To take a simple example, if you’re spending the same amount on social marketing as on paid search but the latter contributes to twice as many conversions, then it’s actually costing you only half as much per conversion. The actual calculation is likely to be much more complex, of course, because there are so many variables including the value of different conversions and the amount of credit assigned to different channels.
Even in the simplest cost model, though, the other complicating factor is taking account of different costs, particularly those that Google Analytics doesn’t automatically know about. It can be a bit cumbersome to import cost data for non-Google search engines and campaigns such as email marketing campaigns and social media advertising, but you won’t get the full picture without it. Briefly, to track, upload and maintain non-Google click and cost data:
Maintain non-Google click and cost data
1. Implement custom campaign URLs – Google’s online URL builder tool makes this reasonably foolproof.
2. Create a custom data set to hold the data from these.
3. Generate the upload data as a CSV file, then import it into Analytics.
4. Report on your data.
You’ll want to establish a routine for updating the CSV file and importing the updates, but without click and cost data for any non-Google search engines and campaigns, Analytics isn’t telling you the whole story.Read previous article Read next article
7.2 Importing data to analytics
How to import data into Google Analytics
Your business probably uses a variety of systems to run different areas of activity. You might have a CRM system to manage and track customer relations, a CMS that includes metadata about the content it serves on your website, and an e-commerce system crammed with product and transaction information. Google Analytics supports the import of data from external sources such as these, enabling you to combine this with data collected via Analytics itself. You can then use Analytics to organize and analyses all this combined data to give you a much more comprehensive picture of your business.
What are the data import requirements?
To use Data Import, you’ll need the following:
- A formatted CSV file: containing the data that you want to upload.
- A Data Set: this holds the uploaded data in Google Analytics and defines how it will be joined with specific hit data already collected by Google Analytics. If you plan to use custom dimensions or metrics as either the key or import dimensions/metrics, you’ll need to create those before creating the Data Set.
- One or more Views: so that you can report on the data.
Once you’ve readied your formatted upload file and target Data Set, you can import the data using the Manage Uploads page (more about this shortly).
What data can be imported to Analytics?
Assuming you’re using Universal Analytics, Google Analytics can import the following types of data:
- User Data: import user metadata, such as a loyalty rating or lifetime customer value, and use these values with Analytics segmentation and remarketing.
- Campaign Data: use campaign tracking IDs and import ad-campaign-related dimensions such as source and medium, letting you expand and reuse your existing non-Google campaign codes.
- Content Data: group content by importing content metadata such as author, date published, and article category.
- Product Data: gain better merchandising insights by importing product metadata such as size, color, style or other product-related dimensions. Note, to import this type of data, you must be using Enhanced Ecommerce (‘Key Account Features of Google Analytics’ playbook, page 45).
- Refund Data: align your internal ecommerce reporting with Google Analytics by importing ecommerce refund data. (This import type also requires you to be using Enhanced Ecommerce.)
- Cost Data: include third-party (non-Google) ad network clicks, costs and impression data to gain a more complete picture of your ad spend.
- Custom Data: provides support for importing custom data sets.
To begin, you sign into your Analytics account, click the Admin tab, navigate to the Property and select “Data Import”. When you configure Data Import, you create a Data Set which defines one or more dimensions to use as a key. Data Import uses this key to match values in the uploaded data to values in your collected hit data. The additional data that you upload is then added to one or more import dimensions or metrics, which can be standard or custom dimensions and metrics. Imported dimensions and metrics can be used in reports, remarketing lists and other Google Analytics tools alongside standard data collected by the website tracking code, mobile SDK or Measurement Protocol.
Note, however, that each import type (as listed above) has a pre-set list of dimensions and metrics that can be used as keys, and import dimensions/metrics as appropriate to that import type. You’ll select from this list of available dimensions and metrics when you create your Data Set. It’s not possible to import using custom variables (these are not supported in Universal Analytics – see ‘Key Account Features of Google Analytics’, page 20), nor can you use time-based dimensions such as hour, minute, etc., nor geo-dimensions like country, city, and so on.
What restrictions apply to data imports?
- Real-Time reporting does not support custom dimensions: any custom dimensions that you’ve imported will not appear in Real-Time reports.
- You may not upload or associated personally identifiable information with User Data Import: for data protection purposes.
- Imported data is subject to any filtering you’ve set up on the Property: this means filters may exclude some data or transform your imported data.
- There’s a limit of 50 uploads per day per Property: also 50 Data Sets in total per Property.
- Maximum size and data limits: there’s a maximum upload file size of 1GB – but then, a CSV file would have to contain a phenomenal amount of data to get anywhere near that size. There’s also a limit of 1TB total data uploaded per Property (Premium accounts).
Once created, a Data Set cannot be deleted. You can, however, unlink a Data Set from all Views to prevent its data from appearing in reports. The uploaded files themselves can be deleted, but this takes effect only from that point onwards. Any reports run after the file has been deleted will not display the uploaded data, but historical reports (from dates prior to the deletion) will still display the data.
Importing and reporting
Once you’ve readied your formatted upload file and target Data Set, you can import the data using the Manage Uploads page for that specific Data Set. This page displays the status of the files that you and others have uploaded for that Data Set. Anyone with Read & Analyze permission can see uploaded files, but you must have Edit permission at the Property level to upload data or delete uploaded files.
Display the Manage Uploads page
1. Sign into your Google Analytics account, and switch to the Admin tab. If you have more than one account, select the appropriate account in the Account column.
2. In the Property column, select the Property that contains the Data Set.
3. Click “Data Import”, locate the Data Set you want, and click “Manage Uploads” at its right.
To upload files, simply click the “Upload File” button and select the files. To delete uploaded files, select the ones you want and then click Delete.
Depending on the type of data that you’ve imported, you can use either a standard report or create a Custom Report to display the imported data. Imported Cost Data, for example, can be analyzed under Acquisition > Cost Analysis, and Product Data under Enhanced Ecommerce > Product Performance. Google provides a handy table listing the import types and available reports for each type.
Note that uploaded data needs to be processed before it can show up in reports. Once processing is complete, it may take up to 24 hours before the imported data begins to be applied to incoming hit data.
To help you get to grips with Data Import, Google offers a set of step-by-step examples of how the process works. Start with the import example overview, and follow up with the set of examples based on real-world use cases, illustrating how Data Import can be used to augment and enhance the default data collected by Google Analytics with offline business data generated by your own systems and processes.
7.3 Setting KPIs
How to set KPI targets and sharing metrics
As we’ve noted, there are so many things you could track in Google Analytics, you need to focus on what’s important, if only to keep things manageable. You might still wish to track a range of metrics for the broad picture, but marketing experts advise concentrating on a handful of the most business-critical measures – that’s what the term “Key Performance Indicator” means. This doesn’t mean other metrics won’t remain relevant and give you valuable insights, but you need to keep your eye on what’s really important to the success of your business.
It’s true, however, that what is “key” may well evolve over time as your business grows and its goals change – and even as the technologies it uses develop. Whether your business has to adapt to seasonal trends or you’re trying to keep up with changes in search engine algorithms, adaption is necessary regardless of what size your company is. It’s also true that there are usually multiple indicators for many goals, particularly if you factor in “micro conversions” or steps along the path to an end goal or “macro conversion”. So, for example, if your business goal is online sales, all the following could be significant KPIs:
- Transactions (completed)
- Average Order Value
- New (or Unique) Visitors to Checkout Page
- Return Visitors to Checkout Page
- Incomplete Transactions/Abandoned Carts
- Bounce Rate for Shopping Cart Page
- Average Pages Per Visit (that lead to transactions)
- Return on Investment (per traffic source)
- Customer Lifetime Value
Crucially, if you’re looking for data on which to base marketing strategy, site development and other business decisions, micro conversions could be more significant indicators than the bottom line – after all, you need to know what you need to improve (such as rate of shopping cart abandonment), not just what is working fine right now.
As we’ve suggested, you’ll often need to consider several metrics in conjunction with each other; otherwise you’ll be risking making business decisions based on only a partial picture. One of the most effective ways to view a whole set of reports at a time is to create a custom Dashboard to give you a high-level overview of KPIs of your choosing. To do this: go to the Reporting tab in your Analytics account, select Dashboards in the sidebar, and click “+ New Dashboard”. You can then select a “Starter Dashboard” prepopulated with some standard widgets to use as a starting point, or “Blank Canvas” to start from scratch.
Finally, when you want to keep others in your business informed and aligned with the same goals and strategies, remember that you can share custom Dashboards and reports in a variety of ways. Dashboards you create are private to you until you share them, but this is easy to do. To share a Dashboard with all users of the current View, assuming you have Edit permission, simply open the Dashboard in question, click Share and select Share Object. You can share the Dashboard configuration (without your data) with other Views and accounts, export the Dashboard with its data as a PDF, send it to others in emails, and share in other ways. The same principles apply to sharing reports and other assets.Read previous article Read next article
7.4 Measuring ROI
How to measure ROI (Return On Investment) with Google Analytics
In this section, we’ve emphasized the importance of applying a comprehensive measure of cost factors. You need to bear in mind that Google Analytics does not do this by default. Look at its AdWords reports for example: once you’ve linked your AdWords and Analytics accounts, click the Reporting tab in Analytics, click Acquisition in the sidebar, and select AdWords > Campaigns. You’ll notice that one of the columns is headed “Return on Investment”. This is simply your total Goal value minus your AdWords costs, divided by those costs. It doesn’t take account in any way of the costs of the sale – in the case of manufactured goods; for example, this would include the costs of materials and manufacturing of all the items sold, in addition to advertising costs. So, at best, although the report column is headed “Return on Investment”, it’s more appropriate to think of it as Return on Ad Spend.
To put it another way, the problem with Return on Investment is the ambiguity of both the terms “return” and “investment” (in this context, costs). In Google Analytics, the investment is simply how much you spend with Google and the return is the total amount of money made by your business or top line revenue generated from advertising spend. In Google’s world, this is appropriate enough: since Google definitively knows only these two inputs for your site, this is all it is responsible for reporting. But neither of these definitions is adequate.
What really matters to a business is profit, and determining profit is a more complex calculation, which may well vary from case to case. Costs include advertising and marketing spend, yes, but also cost of goods sold plus your fixed costs and operating expenses. In the context we’re concerned with here, your fixed costs will need to be distributed across all your marketing activities, so paid search won’t account for 100% of your fixed costs unless it is 100% of your sales. Some fixed costs (such as PPC agency fees) will be 100% attributed to paid search, but others (such as website hosting fees) will not.
Conversely, the “return” side of the equation might also be more complex. The simple definition doesn’t, for example, take any account of the “lifetime value” of users. Customers typically purchase more than once, so if the same user later spends more money with your site – say half as much again – then your return from the same ad spend is 1.5 times greater. This is difficult to incorporate into standard reporting, if only because the time-scale may be significantly greater than your typical reporting period, and there might be many different types of interactions involved in the path to a user’s subsequent conversions.
The lesson, ultimately, is that Google Analytics is a set of powerful tools and not a template for business success. It measures what it measures, but it’s no substitute for your own judgement and business savvy. As marketing consultant Jeff Sauer puts it:
“While it is a thing of beauty to be able to use a single tool for multiple levels of analysis, we also need to know that there is more to running a successful business than minding the reports provided in Google Analytics. This is a lesson that I learned the hard way as I ran an e-commerce business that appeared to be successful in Google Analytics, yet failed to make a profit in real life.” Read previous article