When it comes to improving website conversions and customer experience, two techniques come to mind: A/B testing and session replay. A/B testing enables you to learn what works best for your visitors by experimenting with different variants of the same content and measuring relevant conversions. Session replay, on the other hand, allows you to view recorded behavior of users to understand the problems, issues, and errors that need to be corrected to improve the user experience.
Recently, the InfoTrust optimization team integrated the two together to help a client determine the optimal experience for their users. Session replay gave the client detailed insights into their testing process, so they weren’t only looking at conversion metrics and KPIs of each test variant. Through this combination method, the client has full visibility into how users interacted with the new elements added in the test variants, and can replay sessions to better understand the customer experience.
Google Optimize and SessionCam
Google Optimize and SessionCam are great tools that integrate well with Google Analytics for insights. Even better, we closed the loop by integrating Google Optimize directly with SessionCam—allowing our clients to go into SessionCam and look for session recordings of users in a specific Optimize experiment (and variants).
This integration is done seamlessly; no code is required when the experiment is set up on Google Optimize, and no additional tracking is needed on the SessionCam snippets for the pages where the tests are running. The integration automatically detects a running experiment, parses the experiment and variant information, and sends them to SessionCam to be added to the recording information. Adding new experiments requires no additional configuration. This allowed our client to take their experiments a step further with the ability to see mouse movement, clicks, form input, and scrolling against a specific variant of the page being tested or content being optimized.
Why is this capability incredibly useful for analysis, insights and improving site experience through experimentation? First, let’s dive into how it works.
How Does It Work?
Google Optimize creates a cookie to store experiment data on the browser. We utilized Google Tag Manager to automatically parse the Optimize experiment and variant identifiers from the cookie and pass them to SessionCam when it records the session. This way, whenever you want to see how users interacted with a specific experiment variant, you can just search the experiment information within the SessionCam interface, watch individual recordings, and observe user behavior and interaction with each variant.
When changing the customer experience and experimenting with web content, gathering evidence is crucial; collect insights supported with data on which variant works best before you commit to the change. Additionally, recordings of the “losing variants” can provide teachable moments. See how users struggled with a certain element and how that affected the conversion. These actionable insights will help you avoid similar mistakes in future experiments or web content.
A/B testing is a great way to collect quantitative data from a relatively large number of users. Google Optimize can be administered at scale to provide you with predictions of how the changes you make may affect outcomes. For example, you may learn that 70% of users click on a blue button more than a green version of the same button. However, this type of data won’t tell you the underlying reasons why users click one button more than the other or other elements of the experience that might compete for their attention.
Sometimes interviewing users directly won’t even tell you why they take the actions that they do. For this reason, utilizing a qualitative tool like SessionCam gives you an opportunity to see user behavior in action. By viewing heatmaps and user recordings you can examine how users interact with your site to understand what they choose to do instead of the behavior you intended them to do.
SessionCam has a number of features that make it easy to find pages and features that may be causing some user friction and frustration. For example the SessionCam customer struggle score can quickly guide you to parts of your site that could use some assistance.
Sometimes these issues can be used to develop opportunities for optimization and testing. For example, analytics can tell you that the form below has poor completion rates, but it may be hard to deduce why. Watching session recordings from high-struggle-score pages may reveal that the form requires a corporate email address to subscribe to job-posting updates. Understandably, some people may give pause to using an email address from their current employer to get emails about new job opportunities. With the SessionCam-Google Optimize integration you can uncover experiment ideas and validate your hypotheses by comparing original and variant experiences.
Scroll reach, mouse movement and click heatmaps can also be used to view and compare the effect of changes to key pages. For instance, you can run an experiment to test the effect of adding or removing CTA buttons to webinar listings on a blog list page. Comparing the heat maps of the original and the variant can give you additional insights when statistical insights aren’t quite enough.
While the process of A/B testing has remained essentially the same for decades, powerful integrations like this one between SessionCam and Optimize provide a wealth of new data that enables marketers to make timely data-driven decisions.