Web Analytics – A Consultant’s Perspective:
Like any data source web analytics can only be valuable if the data it generates is an accurate reflection of what it is supposed to be measuring. This is why I am regularly employed by companies to audit their web analytics and improve tag implementation and reporting.
This experience has shown how diverse organisations are at implementing and using web analytics. This means that best practice is often not adhered to and that sometimes fairly small changes to the setup of analytics can have a big impact on the value the organisation obtains from data. In this blog I’ve outlined some of the simple changes you can make to ensure you get valuable insights from your web analytics reporting and analysis.
1. Test that your website and analytics work:
You may be surprised by how few companies test critical user journeys on a frequent basis. As a result the first thing consultants often have to do is to raise bugs and tickets to get the user experience fixed. This is fine if you want to pay contractors to spend their first few weeks fixing your website, but otherwise it is wise to get your own people to audit key journeys before you employ a consultant or web analyst.
2. Plan ahead for web analytics implementation:
Do some research or seek expert advice ahead of implementing or auditing web analytics tracking. This can be as simple as making sure elements on your web site have unique IDs to make event tracking more straightforward and allow for integration with form analytics or UX solutions (e.g. SessionCam). Ensure you can give the implementation team access to a test environment (e.g. dev or UAT) and test accounts that allow them to view every user journey you wish to track.
It’s simple stuff, but it is surprising how if it’s not done ahead of employing a contractor or someone on a permanent basis it will slow them down if they have to wait for these things to be resolved. Getting permissions for systems and environments in larger organisations can be a long process and so get it organised before they arrive.
3. You are not working agile:
Tag managers such as Google Tag Manage (GTM), Tealium and Adobe Launch have revolutionised web analytics allowing more sophisticated and agile web analytics implementation. GTM is a free solution and yet many companies continue to use Google Analytics base code rather than migrating to GTM.
The main benefit of a tag management implementation is that it minimises the amount of tracking and advertising script on you site which can significantly improve the speed of your page load speed. This can literally save you money by improving you conversion rate.
In addition, a tag management platform allows you to be more agile by creating new tags to track user events (e.g. clicks on CTAs) without the need to have a developer hard code it on the web page. Further, when analytics or advertising code needs changing, you only need to update it once, in the tag management platform rather using valuable developer resource to change it on the website.
4. Integrate analytics across separate domains and applications:
Websites often go across different domains, use iframes or have embedded applications that call other websites to serve dynamic pages to the user. These complexities may appear to be a blocker to implementing web analytics across the full user journey. This doesn’t have to be the case.
By simply adding the same tag management script to each element of the journey gives your web analyst the ability to join up the individual sub-journeys into a single and more meaningful user journey. Your tag management platform provides the tools to combine data for cross-domain tracking and applications to ensure you can view user behaviour as if your website didn’t have these common complexities. Don’t let anyone tell you that you can’t track the user journey from beginning to end.
5. Implement test procedures:
You wouldn’t release changes to your website without first testing them in a development or UAT environment, would you? So don’t do it with your web analytics reporting. All this requires is the setting up of separate raw data, test and reporting views in Google Analytics.
This ensures that if a new filter or other change doesn’t work as expected it won’t affect the accuracy of the data you report to management. Such a process safeguards data and allows you to assess the impact of changes to your web analytics before you implement them into your reporting view.
6. Segment your data using separate views:
One of the most useful features of Google Analytics is creating conversion goals and then using the funnel visualization and multi-channel reports to show drop off rates at each step of the conversion journey. However, it’s not possible to segment funnel visualization and multi-channel reports and so to understand how different user segments convert requires separate complex tabular analysis.
To allow you to use these reports for individual customer segments you can create separate views for each user segment that justifies special attention. This also means you don’t have to use advanced segments for every report you run to exclude segments that may not even be in the market for your product or service.
The views you should consider creating include your key target market (e.g. country, region or city), organic only traffic, paid traffic, search engine (organic and paid), social media traffic, mobile traffic, desktop, referral traffic and finally direct traffic. Depending upon the type of business and customers you are analysing you might not need all of these or you may identify others that are more relevant for you.
You should certainly try to exclude traffic that will never convert because you are unable (e.g. for legal reasons) or unwilling (e.g. it’s not economic to do so) to transact with them. Otherwise your conversion metrics will not reflect the reality of your situation and they will be skewed by visitors who will never convert.
7. Google Analytics is not free!
Although you can choose to use the free version of Google Analytics to minimise the cost of web analytics this does not mean you don’t need to spend time and resources on configuring it correctly and ensuring the data is meaningful. Garbage in, garbage out is as relevant to analytics as it is to any form of data collection and analysis.
Make sure you regularly conduct an audit of your web analytics to optimise the quality of data as otherwise you may be drawing invalid conclusions from reports which could cost you millions in lost revenues. Use our Google Analytics audit checklist to complete this process. This includes setting up filters to exclude internal traffic and consultancies that are working for you, bots, referrer sites that send malicious traffic to you and exclude users who are unable to ever purchase from you because of logistical or legal reasons.
8. Audit Google Tag Manager:
GTM also needs careful and planned implementation to optimise data collection and integrity. Use our Google Tag Manager audit checklist to undertake a robust audit process. This includes creating a logical account structure, installation, data layer basics and content security policy adjustments.
9. Implement tracking before a new journey is released:
As all new builds on a website have to be planned and go through a set release process there should be no reason why analytics implementation, event tracking and testing can’t be included in this process. Why wait for a new journey to go live before you start tracking its impact on users and your bottom line? In the offline environment this would be seen as insane. Event tracking can be tested in your dev and staging environments just like any other changes.
10. Document your implementation:
Web analytics and tag management soon gets complicated and even the most loyal employee will move on sooner or later. As your analytics requirements grow and you become more sophisticated in your analysis it is essential that you don’t rely on the imperfect memory of your analyst or contractor.
Make sure you have proper processes in place to document Google Analytics set up and tag management. This will protect you against the loss of expertise and also give new analysts and contractors a good starting point to understand what they are dealing with.
Do some preparation before employing analytics consultants. Audit your website to check that it works and get IT to add unique IDs to elements on each key page of your website.
Similarly try to integrate your web analytics code across all your domains and sub-domains so that your analytics can measure the entire user journey.
Establish test procedures and separate views so that changes to filters and settings don’t risk undermining your management reporting.
Ensure you segment by views so that you don’t have to always rely on complex filters to examine key user groups.
Invest in the setup and configuration of Google Analytics as it is not a free service if you want to get real insights an accurate data from it.
Grant consultants access to development and staging environments so that tagging can be implemented and tested before new user journeys go into production.
Finally, don’t forget to document your web analytic property set up and also your tag management plan so that you allow for staff and contractors moving onto new positions.
If you follow these guidelines you will find you will get more value from web analytics and if you do need to employ consultants they will deliver more value to your business.
- About the author: Neal provides web analytics and CRO consultancy services and has worked in many sectors including financial services, online gaming and retail. He has helped brands such Hastings Direct, Manchester Airport Group Online and Assurant Ltd to improve their digital marketing measurement and performance.
- Neal has had articles published on website optimisation on CXL and Usabilla.com. As an ex-market research and insight manager he also had posts published on the GreenBook Blog research website. If you wish to contact us please send an email to email@example.com. You can follow us on Twitter @conversionupl, see Neal’s LinkedIn profile or connect on Facebook.