Your Custom Text Here
I recently saw people praising a chart from The European Correspondent on social media comparing the years of compulsory schooling for different European countries.
While it's certainly creative, when I tried to actually read the chart and spot the insights that were in the callouts, I had to work pretty hard, and some insights were difficult—or even impossible—to spot. In this blog post I critique the original design and propose a redesign that I think performs better.
Read moreOne of my pet peeves is dates in charts that are ambiguous, but the problem can easily by avoided by choosing good date formats.
Read moreProbably the most common way to show values on a map is the "colored-region" or "choropleth" method. I'm not the first to notice, however, that this method presents some potentially serious perceptual risks.
Read moreI was recently invited to deliver a webinar as part of JMP's great Statistically Speaking series. The webinar consisted of a 35-minute talk entitled "How to choose a chart type (it's trickier than you think)," followed by about 25 minutes of Q&A from a very engaged (and large) audience. The recording is now available to view online.
Read more
I recently logged into the dashboard for my email list management app and noticed that the chart of my subscriber count over time suffered from several common design problems that made it unnecessarily hard to read and potentially misleading. In particular, the quantitative and time scales were less than ideal.
Check out this short (8-minute) video of me fixing these problems step-by-step:
Read more
I recently recorded a video of me turning an “OK” chart into a great chart, step-by-step, to make it easier to interpret and to make key takeaways much more obvious.
Let me know if you agree or disagree with any of my changes!
When it comes to the maximum number of parts that should be shown as a pie chart, there are plenty of opinions floating around out there. In this article, I argue that the maximum number of slices depends on the situation and that, in certain situations, even a 30-slice (!) pie chart can be the best choice.
Read more
An on-demand (pre-recorded) version of my flagship Practical Charts course will be available on June 25th! This self-paced video course consists of 6.5 hours of video in 44 lessons, and is available for 25% off during the pre-sale period (before June 25th).
Read more
When should you use a bar chart instead of another chart type? Easy question, right?
As it turns out, choosing when to use a bar chart or another chart type is surprisingly complex. In fact, I’d argue that, in a way, bar charts are actually the most complex chart type. Sound strange? Then read this article…
Read moreI’ve recently seen a few comments on LinkedIn from people who don’t feel that it *ever* makes sense to use a stacked bar chart to show the breakdown of a single total. While relatively rare, there is a specific situation in which I think it *does* makes sense, though. What is that situation? Find out in my latest article...
Read moreCreating “simple” charts is much harder than it sounds. Telling a chart creator to “keep it simple” is like telling a student driver to “avoid traffic accidents.” Well, yes, obviously, but how, exactly, does one do that? While avoiding accidents sounds simple, it requires learning hundreds of rules of the road and days of training and practice.
Read moreI’ve been asked a number of times how I went about writing Practical Charts because it’s so, well, practical. This is a wonderful compliment to receive and so, as a cheap thank-you, here’s the story behind the book.
Read moreI recently had a great conversation with data visualization researcher and professor Enrico Bertini, about how surprisingly tricky it is to choose a chart type. Watch the video above (20 mins) to discover...
Why both the nature of the data AND the purpose of the chart must be considered when choosing a chart type
Why traditional "chart-type-chooser" diagrams and catalogues are often unhelpful, and even unreliable
Examples of more reliable and helpful chart-type-choosing decision trees from my Practical Charts course
Agree with what we said? Disagree? Great! Let us know in the comments below or in this LinkedIn thread.
You’ve probably seen “smoothed” or “curvy” line charts like this before. Some people like them and others definitely don’t. In this blog post, I explain when I think it’s okay to use smoothed line charts and when they should be avoided.
Read more
You're looking at a chart and it contains a date labeled “10-02-2007.” What day is that, exactly? Feb. 10, 2007? Or October 2, 2007? How about “10-Feb-07”? Is that Feb. 10, 2007, or Feb. 7, 2010? To ensure that your chart readers know what the dates in your chart actually are, choose a date format in which:
The month is alphabetic (e.g., “Aug” instead of “08”)
The year is four digits (e.g., "2011" instead of "11")
As long as the date format that you choose checks both of those boxes, your readers will always interpret the dates in your charts correctly.
Read more
You read that right: MORE Practical Charts! 😉
My first book, Practical Charts, covers about 30 common chart types that are familiar to most audiences, such as bar charts, line charts, and maps. More Practical Charts covers 20 additional, more "advanced" chart types that tend to be unfamiliar to audiences without scientific or technical backgrounds:
Box plots
Merged histograms
Overlapping histograms
Distribution line charts
Distribution heatmaps
Merged bar charts
Scatter plots
Colored-dot scatter plots
Bubble charts
Scatter plot matrices
Pareto charts
Heatmaps
Shape size charts
Overlapping cycles charts
Cyclical heatmaps
Cycle plots
Strip plots
Jittered strip plots
Merged strip plots
Histograms
If you have a technical or scientific background, you might assume that most people know how to read most of these chart types but, after working with thousands of chart creators in hundreds of organizations, I've realized that that's not the case (unfortunately). There's a reason why you rarely see these chart types in the mainstream media: editors know that many readers/viewers won't understand how to read them.
If, however, you do create charts for relatively data-savvy audiences, by all means, pick up a copy of More Practical Charts 🙂. The book will be published on January 16th, but you can pre-order your copy now to get it shipped to you on the publication date. It's relatively short (about 100 pages) and is priced at less than half the list price of Practical Charts.
Why split these 20 chart types out into a separate book? Well, a large fraction of my readers can't use these chart types because their audience would struggle too much to grasp them, or simply wouldn't be able to grasp them. I didn't want to force those chart creators to buy and read content from me that they'll never use in practice, so I packaged that content in a separate, optional companion book.
BTW...
Interested in taking my Practical Charts course and/or my Practical Dashboards course? Registration for my January live, online workshop is now open (but it's already 50% sold out and early-bird prices end January 8th, so don't wait too long if you plan on registering). I hope to see you (online) there!
For decades, chart creators have disagreed on whether pie charts should be used, but I think there are a few important arguments that have been missing from the debate and that might—just might—help to bring it to some kind of conclusion.
Find out what those arguments are in my latest article for Nightingale, the Journal of the Data Visualization Society.
When creating charts for “everyday” reports and presentations, it's generally best to stick to chart types that you know are familiar to the audience. Sometimes, though, you can’t use a familiar chart type, either because there aren’t any familiar charts that can accurately represent the type of data to be shown, or because there aren’t any that can communicate the specific insights that you need to communicate about the data.
In these situations, you might have to use a chart type that you know is unfamiliar to the audience. For example, you might have to use a scatterplot or step chart, even though you suspect (or know) that the audience is unfamiliar with that chart type. What to do?
There are three techniques that I use to quickly teach audiences how to read an unfamiliar chart type:
Gentle reveal
Bait-and-switch
Duh insights
What, exactly, are these techniques? Let’s see some examples, starting with…
Read more
There are six formatting mistakes that I commonly see in pie charts “in the wild.” This article discusses how to recognize and avoid those formatting mistakes in your own pie charts.
Read moreIf you haven’t been using step charts, there’s a good chance that you’ve been misrepresenting at least some of your data to your audience. That’s why I suggest that, if you can, try to start using step charts whenever you need to show irregular, persistent time series values, and start getting your audiences used to seeing this chart type.
Don’t know what a step chart is? Read this post!
Read more