Topics in Data Viz

How We’ve Learned Data Viz, and Why You May Want To Do It Differently

The Data Visualization Society members share how they began learning data viz and what could have made it easier.

Learning data viz. Original illustrations by Martin Telefont.

👋 “How do I learn data viz?”

This is probably the most common newbie question. It’s a hard question to answer because it depends on where you’re starting and who you’re asking.

A couple of years ago, I wanted to learn more about data visualization, so I wandered over to the Technology section in Barnes & Noble and picked up Nathan Yau’s Data Points. I devoured it because it’s an approachable introduction to different chart types and encoding methods.

In the book, Nathan mentioned an “up-and-coming analysis software” called Tableau, so I watched their online tutorials and made some cringe-worthy charts, like this beauty:

Exhibit A: beginner’s work!

I could tell it was a powerful tool, and I really wanted to get better at designing visualizations. So I began reading blog posts about the vizzing process from pros like Nadieh Bremer, but I still wasn’t sure what made a “good” viz and why.

Answering that question happened slowly, through lots of practice. Mistakes and feedback from others helped me to improve over time, but it’s been a long road and I wondered if there was a better way out there.

So, to see how other people were learning, I visualized the 627 responses to these questions from the 2018 Data Visualization Survey:

  1. Did you learn how to do data visualization in school or did you learn how do it on your own?
  2. What method do you think best teaches data visualization?
Viz by Alli Torban using RAWGraphs with data from Elijah Meeks.

It’s clear that most people are self-taught (83% of respondents), and almost half think that examples (43% of respondents) best teach data visualization. If you read through the “Other” answers, many people think that it takes a mix of methods to learn data visualization.

So, the question remains: how do you learn data viz? My experience says through practice, though many others would say through examples.

But to a newbie, that probably sounds like really vague advice. Time to ask the Data Visualization Society

Here’s what we asked the Slack group:

“Did you ever study data visualization in a structured environment? If so, what were the strengths and weaknesses of the courses and programs you were in?

Have you taken any data visualization training? Which approaches work and which ones don’t?

Have you ever taught data visualization? What did you find was effective and what wasn’t?”

Many responded that they hadn’t spent much time learning in a structured environment, but when they did, it was usually an online course (free & paid), one-off workshops, or part of a course in another degree program. Here’s a short summary of the discussion:

What was the MOST effective part of these classes?

  • Learning about gestalt principles.
  • Studying fundamentals and then putting them into practice by creating visualizations.
  • Working through projects that can be added to a portfolio.

What was the LEAST effective part of these classes?

  • Short, online courses that don’t have depth (no statistics or analysis).
  • Working on a visualization before figuring out the story or message.
  • Learning to build very custom visualizations since they tend to be expensive and inefficient.

What’s worked well when teaching data viz?

  • Focus on teaching a design process (e.g., understand your audience, prototype ideas, and test).
  • Use sample data that is relevant to students to make the lessons more engaging and real — better yet, let students choose their own topic and data.
  • Teach critique and give a lot of feedback.
  • Try to follow-up with students after the fact to coach through real-life problems.

A Better Way to Learn

I realized that we often try to learn data viz by fast-forwarding to the “fun” part of visualizing data (i.e. practicing and examples), but that can feel like stumbling through a dark room.

By hearing from data viz practitioners as they look back on their journey, perhaps better advice for learning data viz would be to:

1. Learn the fundamentals.

2. Define your design process.

3. Put it into practice with real-life data.

4. Get lots of feedback.

A great way to do this is by collaborating with a more skilled professional.

For you eagle-eyed readers, you probably noticed that this was the second most popular choice in the survey about the best method to teach data viz. It’s easy advice to look over — I honestly didn’t notice it right away— but it makes sense. Learning is so much more efficient when you see everything in action.

The Takeaway

While practicing was an essential part of my learning, I wonder how much time I could have saved if I had sought out a mentor first, rather than poking around by myself. I could have found out sooner that data visualization is a process rather than the single act of making a chart.

By working with someone more experienced as you learn data visualization, you have the benefit of seeing the full process in action. It also shines a light on what your strengths and weaknesses are so you can further focus on your studies.

If you’re not sure how to connect with a more skilled professional, join the Data Visualization Society, which has a very active Slack group — including channels specifically for newbies and teaming up. They’re also working on implementing a mentorship program. Join us! :)


Thank you to the Data Visualization Society members for sharing your expertise, Martin Telefont for your original illustrations, and editors Alyssa Bell and Jason Forrest for your thoughtful polishing.

Data Visualization Society

The publication for the Data Visualization Society, an initiative to foster community for data visualization professionals of all backgrounds.

Alli Torban

Written by

Data Visualization Designer⚡️Host of Data Viz Today Podcast 🎧 www.dataviztoday.com

Data Visualization Society

The publication for the Data Visualization Society, an initiative to foster community for data visualization professionals of all backgrounds.