Designing Data – Part 3: Look at the Data

May 23rd, 2012 by Alex Carusillo

We’re asked about our design all the time – usually in an incredibly kind way full of high fives and “how’d you do that?!”s but sometimes in a “ugh, did you even think about talking to a customer??” kind of way. So, we decided to give you a week-long deep dive into our design process in this “Designing Data” series. Yesterday we identified problems through user research, and today we'll look at the data to validate our thinking. We’ve identified the problem. Next step is to start sketching or to open up Photoshop and start throwing pixels on the screen, right? Nope. Negative. Wrong.

We always look at the data before even thinking about design.

We’re a company that eats, sleeps, and dreams information and insight, and we design from what we have -- not what we wish we had. It’s so easy to spend days pushing pixels in Photoshop making the perfect design for the perfect scenario, but it’s all meaningless if the result doesn’t carry over to reality. Great mockups are easy. Designs that work well with real data of all kinds are hard. Some days, really hard. When we were designing the Overview on the Chartbeat dashboard, we took each question people were trying to answer and made sure the data we were tracking matched up against those questions to provide real, unique value. We questioned and vetted every decision as to whether or not to put each data point on the Overview, a secondary page, or not include it at all. We looked at client after client, page after page after page trying to find places where our data fell down.
  • People wanted to know where traffic was coming from. But did that mean dropping a list of each referring site was be enough?
  • How did the dashboard look if your site had very low traffic, like 0-10 concurrents?
  • What if it had a ton, like 500,000+?
  • What if everyone was coming from search? Or social?
  • What if you got a flood of traffic all at once? Would the visualization work? Would it tell you that something special was happening?
To some people it can seem like overkill. But when you're working with data, you can't make assumptions and expect the design to match every possible form of data across thousands of sites. Otherwise, it's a ridiculous waste of time. Which is why we've gotten into the habit of  ending meetings the second someone says “Well, if the data says ____” by responding with “Let’s go look at it.” And all gather around my screen, or Isaac's (our chief data scientist's), or Matt's (our UI designer). We tear apart every single concept that makes it to this second stage in the process. When they survive this grueling data-review round, the fun part begins: Designing and prototyping - which we’ll dive into tomorrow.