Analytics Design vs. Interaction Design

Interaction Design is easy for me. I can see the big picture and architect the system with all of the right patterns. When it comes to Analytics Design, I find my inner muse drags its feet. I am not talking about designing a BI tool like Tableau or Good Data. These have some interesting interactions. I am talking about the Analytics section of a workflow application.

In my mind, there are two types of people. Dashboard people and analytical people.

Dashboard People
This is the majority of people. They don’t want to slice and dice and explore. They want to know what is important without alot of work. I have heard the sentence a hundred times, “We just want a dashboard!”. This sounds good, but the problem is that dashboards typically do not get looked at with any frequency. Plus, the information on a dashboard is typically not actionable.

So this creates a tension. I can design something stupid that checks the box or something useful that people won’t use.

Analytics People
I made the analogy that Analytics is like Jazz. It’s really cool, but most people don’t understand or appreciate it. Truth, real truth, can be found inside strong analytics. The problem is that it takes discipline and knowledge to get there.

Take Google Analytics as an example. 95% of all Google Analytics users do not make custom charts. They just use the built in ones. However, if you really want to learn about your site, you need to learn about segments and filters. (Great site about analytics)

The funny thing is that dashboard people and analytics people have wildly different views on what data is and what it’s used for. It’s difficult to build a single system to solve the problem.

Well, enough procrastination. I gotta get back to designing this dashboard.

Leave a Reply