A tool for finding what makes your customers unique

Hero Image For Analytics

Nº 01Overview

TrueaData’s Analytics show what makes an audience unique. For example our analytics could show ACME CO.’s demographics are made of 85% people under the age of 35; pointing to a marketing plan designed for millennials. I modernized our analytics to enable marketers to make clear and decisive actions.

When I joined TrueData, this is what our analytics looked like:

Nº 03Improve Data Sources

My first step whenever I begin designing is usually to talk to our customers. For this project it was more important to go to our Engineers/SMOs to learn about what was feasible, and ask two important questions:

"What data do we already have that we aren't using?"
"What data do our partners have that we havent bought yet?"

I gathered 54 new data types that we weren't currently showing our customers, and then went to go and talk with them to begin learning. I did a quick poll and card sorting activity to find out what was most useful from the data that we currently displayed, and what (if anything) was not useful.

Next I repeated the exercise, this time adding in the new data sets. We learned a ton about what our customers were most excited about having access to, and what was essentially useless.

Finally, we had a tehcnical and buisness model discussion, to maximize value and the time of our limited engineering resources.

Nº 04Update Layout

As we added more analytics I implemented a grouping system, so that data types could be compared and sorted by tabs. With a long list of available graphs, and each graph being a vertical bar chart, the layout was too homogenous and prone to losing your place in.

Nº 05Add Color

One of our non-measurable goals was to get our customers to have our information running in the back of their head after they leave our platform, and inform their marketing strategy. My testing showed that our customers were having a hard time differentiating inforamtion quickly, and retaining that information for long periods of time(3-7days). The simpliest way to improve this is by increasing levels of redundant encoding (adding multiple differentiators to data vizuals). So along with our existing labels, and newly added groupings, we added a robust color library, to seperate charts in a multitude of situations.

Nº 06Add Charting Options

Next, we wanted to give our customers multiple ways to vizualize the data, and added multiple new charting options. From everything being unvarying vertical bar-charts, to adding: Stacked Donut Charts, Convex Tree Maps, Angular gages, Packed Circle Charts, Trendlines, and more.

Nº 07Make Data Actionable

Finally, and most importantly, it was critical that our customers be able to actual;ly act on the data in front of them. Not open in a new tab, not take our data to a competitor, but click on a bar on a chart, and start building an audience from it.

In the same line of thought, being able to directly compare two (or more) audience was a new and heavily request feature.