After months of data analysis and research, the results of the 2019 Design Census are here. Some of the information confirms what we suspected, and some of it is kind of surprising. Taken together, it tells a compelling story of where the design industry is right now, and it gives us a clear sense of what we need to work towards in the years to come. 

This year’s census questions are more expansive and inclusive than ever before, and we used that added context to help paint a more accurate and nuanced picture of the state of design. You can see the top highlights and interactive data visualizations on designcensus.org, where you can also download the full 2019 Design Census Report PDF for a deeper dive into the data. We also encourage designers to download the raw data and make their own visualizations (more info on how to do that, below).

But before you dig into the numbers, test your assumptions about your fellow designers and take our quiz to see how well you already know the design industry:

If you’ve got a thing for data, we want to help you parse the responses to the 2019 Design Census with some methodology tips from Archie Bagnall, former chapter president of AIGA Orange County (who previously sifted through the 2017 Design Census raw data and created this very accessible article):

  1. Name any data ranges that you’ll be working with (i.e. “CA,” or “NY”).
  2. Create dropdown lists to help filter the data down to your specific interests (this YouTube tutorial is also helpful).
  3. Pre-build basic charts for questions that you’ll be asking regularly (i.e. Salary vs. State), that can be updated via dropdowns.

If you’re feeling more ambitious and want to be a hero, you can also format the data into tables so other designers can stay out of the raw data altogether (see how Bagnall did this for the 2016 Design Census).

Note: as you play with the data, the accuracy can vary wildly depending on how it’s filtered, and it can be easy to accidentally (or intentionally) manipulate the data and draw misleading or inaccurate conclusions. If you’re not a data scientist or statistician, there’s not an easy way to get an exact measure, but comparing your filtered results against the dataset as a whole can be a helpful barometer. Bagnall strongly encourages table formatting as a safeguard.