Lately I’ve added CrazyEgg to my quantitative toolkit. CrazyEgg is a scroll mapping and click mapping tool that produces time-based snapshots of any static page that has a tracking code injected into it. it has helped contextualize user behavior with some pretty elegant visuals. While it isn’t as precise as an eye-tracking study and the tool itself has some issues, on the whole, it’s great. Here are some of the benefits, issues, and applications I’ve come across in a couple of weeks of daily use.
First, benefits:
- It’s incredibly simple to use. Create a method of injecting the javascript package onto every page of your site and you immediately have the ability to deploy snapshots of user behavior into any static (or relatively static) page.
- It provides a great range of visuals. Scroll mapping, click mapping (with integrated data like referrals, time on page, etc.), and heat mapping.
- It has a simple to use API that should lend itself really well to automation down the road.
Now, some issues:
- While it has excellent and simple scroll-tracking, the click mapping implementation doesn’t have support for many common UI patterns. For example, it won’t play nice with flyouts and floating UI. A huge improvement would be some way to connect animated elements to snapshot states.
- The new version of the website is pretty clearly a re-platforming that didn’t quite get completed. Dead links, confusing UI and documentation, or even occasionally wandering back into CrazyEgg v1.0 aren’t uncommon experiences.
So… what are the best applications thus far?
When I’ve combined CrazyEgg with other digital analytics tools, particularly other event tracking tools, it provides great indicators about what the user is focused on even without an eye-tracking study or full mouse-tracking suite. It’s surprisingly lightweight. Blending the three core views together (scroll, heat, and click) gives a great picture of where the action is on the page vs. where common UI patterns are. Looking down the road, one of the really key applications could be an implementation where a custom developed layer sits on top of the CrazyEgg API and alerts the UX team to emerging patterns of behavior, rather than relying on hands-on analysis.
Overall recommendation? Definitely worth the money for the basic package.
