Human-centered data visualization

Evelyn talked to us about human-centered data visualization. In the podcast episode, you'll learn all about data visualizations that really help people.

Here you can listen to the episode

Are data visualizations always human-centric?

Evelyn, visualization expert for data products, clearly says: No! Actually, it's like with every digital product nowadays: A data visualization must be designed for the users behind it and focus on the people. But that's usually easier said than done.

Why is it simply more difficult to design data visualizations in a human-centric way than other products like apps? Data is predominantly such a big topic and its generation is so complex and costly that users are completely left out. On top of that, the people who collect data are largely not the users in the end.

Furthermore, 95% of data visualizations are not designed by designers. Data analysts do not see it as a matter of course to interview users, to research user needs and to generally drive a design process. This leads to dashboards being built that are not used in organizations because they don't really support the users. This leads to analyst burnout: data analysts are constantly building new dashboards and data visualizations with no added value and no real improvements. The needs of the users seem to change constantly.

In addition, data represents a language of its own through which certain facts are expressed. Not everyone is able to understand and translate this language.

Data in the real world

The nuts and bolts of human-centered data visualization is its inventory in the real world. Designers must ask themselves the question: What real-world problem do you want the data to solve or help solve? Let's say a data analyst is collecting data about marketing campaigns. The goal of the using marketer is to compare campaigns. But what exactly is the goal? For example, the goal might be for the marketing expert to infer whether the new campaign was better than previous campaigns. The data visualization should be structured to answer exactly this question quickly and easily.

In the end, data visualization must help users make a decision.

Best Practices and Other Recipes for Disaster

There are a lot of things you can do wrong with data visualizations. One example is best practices. These only add value in individual cases, but understanding users is more important than any best practices. Of course, there are some small things that are (almost) always valid, like not trying to pack multiple pieces of information into one chart or writing the question the chart answers in the headline. For more such valuable tips and Recipes for Disaster, check out the latest podcast episode! We'll preface the best tip here: if you're unhappy with your dashboards, really try to get behind the question that needs to be answered. Ask questions until you really understand what problem the data visualization is supposed to solve!

As always, feel free to send your feedback to:

Here you can find Evelyn on LinkedIn:

Here you can find Evelyn's Chart Doctor Data Product School:

Written by: