If your data involves multiple units of time, like years / quarters / months / weeks / you see where I’m heading with this, you’ve got a story about change over time.
What’s the best way to visualize that data?
My friend, you have so many options.
The Standby
The easiest, most common go-to chart for showing change over time is the Line Graph.
With most line graphs, time goes along the horizontal axis (aka the x-axis). Oldest years would be on the left and more current years on the right, when reporting to cultures that have a left-to-right orientation (otherwise, check locally about what’s convention).
Each metric you’re reporting gets its own line.
If you don’t know what else to do with your data, start with the Standby. It’ll be a decent backup option, just like how every great breakfast joint will always offer the good ol eggs, potatoes, and toast.
While this is the standby choice for change over time data, you actually have a TON of other options that can highlight different angles of your story.
For good reason, too. It’s easy for a line graph to get tangled with you have lots of metrics with similar values, like this, The Spaghetti:
The Corporate Type
I’ve nicknamed this chart The Corporate Type because it was popularized by the McKinsey Consulting Group inside corporate America. For a long time, only people in that world really understood how to read this chart but then it began to trickle out into our newspapers and blogs and gained even more popularity.
The Waterfall Chart.
It emphasizes the increases and decreases by making them thick columns (rather than relatively more subtle changes in direction that you’d see on a line graph).
What I also love about waterfall charts is that they don’t necessarily have to represent equal increments of time along the x-axis, the way that would usually be required in a line chart. Waterfalls are suitable for showing the steps of a process of sequence, in chronological order.
The Dynamic Duo
Whenever I’ve got just one metric over time, I feel like it looks a little lonely. Just one line, suspended in the air, all by itself. Not much to look at, you know?
So I like turn that particular data into a Line + Area Graph, where the space under the line is shaded in via an area chart.
There’s just more for your eyeballs to grab onto.
The Editor
Sometimes all the up and down – all the change in your change over time – adds unnecessary noise to your visual and the conversations that stem from it.
That blip in 2020? To be expected. Don’t let it take away from the big picture of the overall trend.
But those blips can be hard to ignore.
Enter the Slope Graph.
It’s change just between two points in time, without any regard for what else might have happened in the interim.
Because all the interim noise is gone, slopes highlight increase and decrease. Stories about how one thing went up when all others went down. Or even stories about how particular metrics stayed the same – the line is flat.
The Wild Child
Hold on to your butts. Sankey Graphs can get intense. They show flow, between many inputs and many shared outputs. Usually, but not necessarily, this implies change over time.
The width of the flow corresponds to the value it’s representing.
But Sankeys don’t have to be complicated.
Here’s a super simple example based on work with one of my clients, who ran a school that had a particular specialty hub for STEM (Science, Tech, Engineering, and Math). They wanted to track, of the students who were part of the STEM hub in high school, who went on to declare a science major in college. And who ended up declaring a STEM major that wasn’t in the STEM hub in high school.
This simple example packs in a lot of interesting data, and a Sankey graph is the best way to show it. Can you see how it’s still a story about change over time?
The Underdog
Radial charts have long been derided among the dataviz crowd. The shape is meaningless, we said. Just unwrap it and make it a line chart, we said.
I thought the same until I connected with specific Indigenous folks who saw time as more circular.
So I set my Western thinking in the backseat and I introduced radial charts as an option in my workshops with various tribal epidemiology groups. Every time, people in those audiences would remark on how much this so-hated chart just clicked their time-based data into place for them.
Radials made sense.
Ok, I’ve given you six different options here… and there are even more than this.
Yay!
Yay?
Maybe having too many options is a bad thing? Sure, it can be.
But if you learn when each of these options works best, you’ll adopt the strategic thinking necessary to pinpoint the proper choice without a second thought.
I’m teaching the nuances you need to know in a free class happening Friday (so act quick, yeah?).
You’re invited to be a guest in the virtual studio audience for your new favorite game show
Which Viz Is It?
Your esteemed host (ahem, me) will showcase 13 different graphs and you’ll weigh in on whether that’s the right chart choice… or not.
Rank up the leaderboard with each correct answer. Even wrong answers will lead to new insights about your chart choices.
Along the way, I’ll explain what works, what would work even better, and – most importantly – why.
I’ll show you how the students in my Data Visualization Academy use our Which Viz Quiz to navigate the nuances and land on the best chart for their data. (Enrollment in the Academy opens tomorrow. Doors close next Wednesday.)
This free class is when you learn the refined strategic thinking about communicating data effectively.
WHEN Friday October 6 at noon Eastern Convert to your time zone here Plan for 60 minutes.
HOW Register below and I’ll wing the login details to your inbox.
WILL IT BE RECORDED Yes. I’ll make the recording available for a few days. But this is a game show. It’s always more fun to be in the live studio audience than to watch it on the couch later.
You need to be able to articulate why THIS chart versus THAT chart, just like how I’ve explained the differences between six ways to show change over time.
You’ll learn about:
✨ Chart types you probably haven’t seen before
🙈 When they don’t work
😍 When they definitely do
💤 What you probably would have picked instead
💅🏽 How new charts make your team look so damn good
And I promise a showcase showdown worth of fun along the way. 🎉
