Your favorite chart type is more flexible than you think

Hi, this is Aya from the support team at Datawrapper. You can usually find me answering questions at support@datawrapper.de, updating the Academy, or setting up your custom design themes.

A question I often encounter in the data vis community: What’s your favorite chart type? Instead of just asking myself that question, I decided to look at all the data from Datawrapper’s past eight years. What charts have been popular amongst our users?

This is a chart type that's commonly known as a bump chart or rank chart. Bump charts (or rank charts) are simple and communicate one thing very well — the rankings and how they changed over time.

As Lisa explained in her annual roundups in 2019 and 2020, we've seen a rise in popularity for our maps after welcoming new members to the team and making improvements, including our recently renewed data upload step. That's clearly communicated with a bump chart.

"But wait," you might say, "I've never seen the option to create a bump chart in Datawrapper!"

That's because bump charts can be created using a simple Datawrapper line chart. Line charts in Datawrapper are very flexible — they can be used to create a bump chart, a lollipop chart, and many more. The lines between the categories are blurry, which might explain why line charts have ranked #1 in popularity in recent years.

Here's an example of how blurry the lines can get.

Two ways to make a bump chart

Here are two almost identical charts: one is a line chart, the other a scatter plot. Can you guess which is which?

Created withLine chartScatter plot
Data formatWide[1]Long[1]
ProcessEasy peasyMight take a bit of time
TooltipNot so customizableVery customizable
LinesEasy to customize, interactiveDifficult to customize, not interactive

Using a line chart

So, how do you choose which chart type to use? Well, it's a lot easier and more intuitive to create one using a line chart. Line charts also give you the ability to curve the lines, like in the first chart of this blog post. This makes it easier for the eye to follow the lines across the page. You can also interact with the lines by hovering over them. So if ranking is all you want to communicate, use a line chart type to create your bump chart.

Bump charts are great because they're so simple. But they also have downsides — especially when you want to show not just the ranking but exactly how popular each chart is.

The Flerlage Twins explore this problem in their article on how reducing data to rankings can sometimes be misleading. For example, in the line chart above, the reader might wrongly assume that people are creating fewer column charts as time goes by. But that's simply not true. The number of published charts has been growing for all chart types, especially in 2020. Even though locator maps stayed the same in ranking from 2019 to 2020, our users created 60% more locator maps in 2020.

This is where a scatter plot can come in handy.

Using a scatter plot

By using a scatter plot, I can display the (relative) number of published charts of each type, as well as the rankings. There is a lot more customization you can do as well, especially in the tooltips. Hover over the symbols to see for yourself!

The caveat: Scatter plots can be a little bit tricker than a simple line chart. You'll have to reorganize your dataset from a wide format to a long format, and the lines aren't interactive either. For a quicker, easier bump chart, the line chart type is the way to go.

Chart types are flexible

I hope this shows just how flexible Datawrapper charts can be, especially line charts and scatter plots. You can see that's no secret in our office — around 50% of all Weekly Charts created in the history of Datawrapper have used either line charts or scatter plots.

At Datawrapper, there are 19 chart type options to choose from in Step 3: Visualize, which might not seem like a lot. There is a reason why we keep the options limited and simple[2]. But there are also a ton of chart types you can create out of these standard 19. I'm excited to see you come up with your own!


If you've read this far, thank you! If you have any questions or feedback, I would love to hear from you at aya@datawrapper.de or on Twitter @ayatnkw. We'll see you next week!

More reading

  • This is not the first time a bump chart has appeared in our Weekly Chart series. Read this post to see a beautiful bump chart created by Roberto Rocha from CBC news.
  1. The difference between wide and long format: Academy article: How to get data in the right format with Pivot tables! ↩︎
  2. In defense of simple charts — Simple visualization types don’t need to be boring ↩︎

Comments