My colleague, Nirish and I ran a workshop, ‘Where UX and Analytics Meet’ for the UXPA's Winter Workshop Extravaganza. I have to say it was a challenge to try and convert hardened sceptics among the 30-strong crowd to the value of analytics in the user experience design process. Analytics and user experience are typically viewed as being very different from one another. The traditional view is that analytics is left brain, user experience (UX) is right.
On the surface they seem very contradictory. However, I believe the truth is far from it, UX and web analytics can and should work harmoniously together.
Any UX consultant worth their salt knows what a persona is, but have you tried using them to segment your traffic in Google Analytics? It might seem like an unusual idea, but it can produce great results! From your personas pick out defining features and use them to build a segment in Google Analytics. For example, you could build a segment based on a persona of someone that...
All of the above are stats that Google Analytics has, or can have, on your users. You might need to enable certain reports to get all this data. While Google Analytics wont have this level of data for everyone that uses your website, it will for a significant enough proportion to make it a worthwhile exercise. The real benefit of creating persona based segments is that it humanises your data, making it much more digestible. It's also practical, as it lets you match website usage data to research based qualitative personas so you can see how different groups of users interact with you online.
Another instance where I find analytics can work harmoniously with user experience is in helping you understand what users are doing on your website, before exploring why they do it. For example, your online sales have been steadily dropping for months, but you don't know where the problem lies. Using Google Analytics' behaviour flows tool (image below) you can see user journeys through your site, and identify problem pages. A problem page might be one where a large percentage of people are dropping off (leaving your website).
Google Analytics has removed the ambiguity of diagnosing the problem. It's showed you what users are doing on your website - in this case, people on the user journey of making a purchase and where they are dropping off. At this point you've narrowed the problem down to a page, form or whole user journey. Now UX comes into its own and tells you why there's a problem, so you can work on designing a solution. A UX tool for understanding why there's a problem might be user testing, for example. This is an over simplified scenario, but illustrates how UX and analytics can work together as part of the same process.
One of the first steps in a project of any sort will be setting goals. A UX consultant will work on developing user goals, one of which might be to enable a user to find a product or service quickly and easily. A web analyst's goal might be to shorten the length of a user flow. These goals are not so dissimilar, in both cases the reason for the them is to make life easier for the user. I'd recommend that at the start of a project UX consultants and web analysts meet to discuss goals and success metrics together. By the end of the workshop I think we softened even the staunchest of critics, and helped the group leave with an understanding of how UX and analytics can compliment one another. All-in-all it was a great evening, and we look forward to running another workshop with the UXPA. If you have stories of working with UX and big data please share them in the comments. Alternatively if you want to learn more about interpreting analytics, I run a training course on the topic.