Every. single. part. of a visual will be interpreted and assigned meaning. Whether you like it or not. Which means we’d better get thoughtful about design.
Myth: The Data Speaks for Itself
The myth: Your job is to design a great study, analyze the data, and then share it. Preferably in a journal article. That others will have to pay to access.
Exploration vs Explanation
Explanation is where you tell the story in the data you saw during exploration to other people who are not explorers.
Fonts for Graphs
When it comes to your graphs, you need condensed fonts.
Must Know Qualitative Charts
Wanna learn about my favorites? Adding these visuals to your knowledge bank will give you new ways to tell stories and get people engaged with your data.
Quant and Qual
Tell me if this sounds familiar.
Back when I was at the university, doing research full time, we’d produce reports where the front contained all our quantitative results and the back held our qualitative findings. We coulda been talking about the same themes in both sections but it was up to the reader to connect those dots.
Why do we do this to our audience?
The answer is usually that there’s a qual team at the university and a quant team and never shall the two meet. We make silos. Like those ex-best friends at the party who just happen to not be in the same room at the same time.
I just wanna Freaky Friday the quant and the qual folks so they could build some cross-team understanding that would make it possible to generate integrated insights.
Because that weirdly contentious divide on the research team creates situations where yall haven’t done your job. You’ve maybe delivered data. But not insights. Information. But not answers to people’s questions.
If the audience has to do all the work to pull out the meaning from related content delivered in two different locations, you’ve failed them. And everyone who contributed data to your study. And me. Because I know you’re better than that.
So let me start you off with the easiest, most low hanging fruit-est way to integrate your quant and qual data.
Place pull quotes near their associated quant charts.
Instead of putting a string of quotes in Chapter 8, pull out one excellent quote, pop it on a rectangle, and slide it right up next to your graph. The combo works together to tell a fuller picture.
Got more than one good quote? Cool. Use them.
In this example, we created a heat map of interviewee responses, color-coded along a Likert-like scale. Quant-y. And we illuminated each Likert-like sentiment with a color-coded corresponding quote.
This, by the way, was the executive summary of the report. Yes, the report audience loved it and yes our client sees us as total rockstars.
When we bring stories and data together, we create the strongest possible chain of evidence. So please don’t let archaic academic divides keep you from producing the best data reporting you can.
Pull quotes are just the beginning. Like your gateway drug to qualitative data visualization. Check out even more qual viz ideas here.
I can show you how to make these stunners (in Excel, Tableau, and R) inside the Evergreen Data Visualization Academy. Enrollment is closed right now, but get on the VIP list and I’ll send you possible discounts and even let you in early next time we open the doors.
So Your Viz Flopped…
When I was in middle school, a cool girl invited me to her birthday party at the country club. I wanted to impress, so bad. This was not my place. These were not my people. But I was gonna try.
So, after luscious cupcakes, I climbed the ladder at the country club pool all the way up to the high dive. I’m scared of heights but my desire to carry out my mission and make friends overruled my shaky legs. Once I got to the top, I realized, I don’t really know how to dive.
So my head told my body to just do what it looks like in the movies: elegant swan dive.
In reality, I bellyflopped so hard.
Some of my hopeful friends snickered. Some just turned their heads away from the crime scene and didn’t speak to me again.
It’s felt the same when my graph and I bellyflopped into the deep end, right down to the emotional damage and painful abdomen.
I’ve learned that if your viz flopped, it’s because you’ve messed up one (or more!) of these four areas.
The point isn’t clear.
Nobody wants to expend their limited time and focus trying to decode a cloudy graph. You’ve got about 3 seconds to get your initial point across. The easiest way to save your graph from flopping in this particular area is to use some text (like your graph’s title) to tell people your point.
See how this works?
Even if you still need to fix up your formatting or rethink your chart choice, your point will be clear.
You mismatched the viz and the audience.
Graphs flop when they aren’t answering the right questions.
Like, your leadership wants to know if the new infant health program is saving lives but you’re in the meeting with a graph about the number of brochures you distributed. Honey, that’s like bringing candy corn to a dessert bakeoff with Martha Stewart. You’re gonna be eliminated in the first round.
We get into this situation when we don’t understand our audience. Or when we simply don’t have data collection set up well enough to track the important indicators that’ll answer their questions.
You can also mismatch when you have the right data to answer the right questions, but the graph is pitched at a different level of data literacy.
Let’s say you’re headed out to staff the booth at a local community festival, where you’ll hope to engage people in conversations about maternal and infant mortality. But you’ve got this as your visual aid.
This visual is pitched for a highly academic audience. (Though I suspect many folks who say they have high data literacy would still struggle to interpret this.)
You’re gonna get people at that festival trying not to make eye contact with you. Because this visual isn’t hooking them in.
It’s aesthetically difficult.
My dear that’s another way of saying your graph is ugly. Oh you thought beauty standards only applied to humans? Nope, they’ve come for your graphs, too.
Thankfully, the beauty standards for graphs aren’t as oppressive and unrealistic as those foisted upon women.
You’ll get halfway there just by changing up the default formatting settings in your software (no matter which software you use).
I can walk you through the formatting steps you need to take. Get my data viz checklist and I’ll point you in the right direction.
You chose the wrong chart type.
Even if you’ve got a clear point in a well-designed graph pitched to the right audience, you can still bellyflop if you’ve chosen the wrong chart type. Your audience is going to read your point in your title and then look to the chart for evidence that confirms your point. If the evidence isn’t obvious because your chart choice is disguising it, you’re done.
And I feel for ya, because most people don’t know about all the possible chart choices that are out there. So we make bad decisions because we don’t realize that a better graph is possible.
I have so much more to say about choosing the right chart type. We teach you the strategic thinking you need to ace this (and how to make the right chart you land on) inside the Evergreen Data Visualization Academy.
Get on the VIP list for possible discounts and early access next time we open enrollment.
The 3 Second Rule
The 3 Second Rule is a major orienting principle that guides the way we make graphs at Evergreen Data. Because our goal around here is to make graphs that are so freakin good, people quickly grasp what we’re laying down and they want more.
You probably have that goal too.
This is not about when food falls on the floor, how long you can do something in basketball, or how much space to leave between you and the car in front of you.
The 3 Second Rule was taught to me by my graphic design friend, Peter. He told me that you get about 3 seconds to grab someone’s attention. That doesn’t seem like much time. But if your design is effective, it’s more than enough.
I’m not saying people have the attention of a goldfish. So insulting. I AM saying that every one of us has a world full of demands that compete for our time and attention.
If we want our work to actually register with people, it has to cut through the noise.
It has to pass a 3 second test.
Cause here’s what happens: People glance at your graph and immediately start running questions through their head like:
What is this talking about?
What’s the point?
Is this relevant to me?
Why should I care?
What’s in it for me?
No one’s trying to be selfish, it’s just that we have to make judicious choices about where we put our limited time and energy. We’re going to spend our resources on the things that seem personal, relevant, and important.
If the answers to those questions aren’t clear inside the 3 second window, people quit engaging with us.
This graph doesn’t pass the 3 Second Rule:
Conversely, when people can answer those questions within 3 seconds, they’re hooked. They’ll come along. Read more. Look at more graphs. Think you’re a complete and total rockstar.
This one has a much better shot of passing the 3 Second Rule:
The 3 Second Rule is not for everyone.
Some people want to make data visualizations that are so intricate and detailed, they look like art.
Gorgeous. Beautiful. Def not gonna pass a 3 Second Rule.
Some people want to create vizzes that invite exploration.
This is Federica Fragapane‘s data visualization on women in politics in countries around the world.
Also cool and pretty – there’s a place in data viz for all kinds of designs.
I just want you to think about what’ll be appropriate for your scenarios and what your decision-makers wanna see in the board meeting.
The 3 Second Rule is for those of us designing data visualizations that need to inform and spur action. The kind that make people stop scrolling and LOOK. The graphics that may not win design awards but make you look as smart and clear as you really are.
Ready to learn how to make charts that follow the 3 Second Rule? Check out the Evergreen Data Visualization Academy. It’ll change your life.
Questions to Ask When Examining a Graph
Does the story told in the headline match the data used to bolster it?
It’s really common for people to read a tweet and take it as the truth. Especially when it comes from official accounts like the White House. And while the Trump White House was notorious for it’s uptick in data viz – that was terrible – the Biden White House data viz is not much better.
Take a look at this tweet:
In recorded history? Holy cow! We are ROCKING IT.
But take a look at the evidence being used to support this claim. The chart’s y-axis begins at Jan 16 2021. That’s just the start of the Biden administration. Not all of recorded history. These two things don’t match.
It’s ok to ask for more evidence.
I hate that we’re here but these days you have to also ask:
Does this data appear accurate and true?
Accuracy (in part) means you pay attention to the scale – is this scale appropriate for this data?
Let’s look at a few examples.
It’s hard to see the problem at first glance. My college-educated intellectual partner couldn’t see the problem until I pointed it out.
The y-axis starts off in increments of 1 and then, conveniently, changes to increments of 0.5 just as we clear the rest of the data and get into the value for 2021. The change in scale at that point makes the 2021 bar look taller than it actually is, as though the GDP grew more in 2021 than reality. (They fixed it here, where the 2021 bar is clearly shorter.)
The scale question shows up in the line chart we looked at earlier… and in a reply the White House posted, in which they pump up the deficit reduction in a column chart.
Looks like a pretty serious reduction, right? Like a third. Wow! Oh wait. The scale starts at 2,000 billion dollars, which is a weird way of saying 2 trillion bucks. The truncated scale here makes it look like the reduction is a LOT bigger than it actually is.
That does NOT always mean that scale must begin at 0 or stretch to 100%.
If 100% isn’t even in the realm of possibility, it doesn’t make sense to include it. For example, let’s say we’re looking at a school’s truancy rate. Since most kids go to school every day, a bad truancy rate would be something like 10%. In fact, an increase from 8% to 10% would be big, bad, and requiring some action. But that increase wouldn’t be detectable if the scale ran to 100%. So, look at what scale would be meaningful for this data. (Sometimes it can be hard to know.)
What data has been left out of this picture?
This tweet contains a data visualization. You can really only tell if you look at it for a while and eventually notice the little legend in the lower right that says one part of a bridge equals 100 actual bridges.
I’m pretty sure I’m gonna have to whip out a tape measure to do this math. But that’s not even my point here. Let’s focus on the claim they’re making: 1,500 bridges, fixed soon. Cool. Wait. Is that a lot of bridges? How many bridges are in the US? Or in my state?
We don’t know the denominator here, so it’s impossible to determine if 1,500 is good or if they’ve set their goals incredibly low. We don’t have the whole story.
Here’s another line chart:
This time about retail sales. Are you questioning whether that scale should start at 0? Me too. Is $0 in retail sales ever EVER going to be a reality? Not in your wildest hippie dreams.
Beyond that scale issue, take a look at the x-axis. It starts in Jan 2020 – which makes it match the tweet where they reference “over the last year.” Good! However, what did retail sales look like before the pandemic? Those data have been left out of this graph, isolating the story to something brag-able. The data that’s missing would paint a different picture.
Building your BS Detector
Every data designer is also an editor, making choices about what data goes in and what stays out. That’s not to say every graph is manipulative – it’s just that there’s only so much time, space, and attention span. So we have to edit. But yeah, it can cross the line into manipulation to support a specific agenda. Your job is to watch for when the line is being crossed.
Be on the lookout for source information – and sometimes look up the source too. Check who generated this data being visualized and whether they did so in way that was consistent and research-based.
Which makes it seem like you have to have a PhD to determine whether a graph is any good. What hope does my grandma have (bless her) when she’s scrolling Facebook? It can be so hard to judge truth and accuracy if you don’t have a background in stats and data visualization.
You’ll get wiser at spotting graph accuracy when you create more data viz in your daily life. You’ll learn what choices have to be made in the process of making this sausage and you’ll be able to see those decisions reflected in others’ graphs, too. It’ll put you in a much stronger place to intervene (with grace) when you see misinformation online.
When to Viz
One of the biggest barriers to making great data visualization is the time it takes to do so. But that’s typically because we wait to start visualizing until it’s late in the game. This post is on when to viz.
The Typical Data Visualization Process
Let’s say you want to run a survey to find out how much employees have integrated what they learned from your company’s half-year intensive training on diversity, equity, and inclusion.
Typical projects like this start off with a topic to explore and perhaps some specific questions that need answers.
Then we convene committees to re-word the specific questions and look at the topic through other angles and eat bagels and yammer on (not a big fan of committees over here). This part of the process takes weeks.
Finally, there’s enough agreement that someone with expertise drafts a survey questionnaire. Someone identifies the pool of potential participants. Someone drafts lovely emails gently nudging employees to take the survey.
The data collection period takes a month or more, because low response rates prompted more committee meetings where consensus landed on extending the survey period and offering incentives for survey participation and someone had to run out for that iPad.
Once the committee has decided enough data has been collected, someone spends a week or more cleaning and organizing the dataset. It was only supposed to take a couple days but the dataset was messier than expected.
Then comes analysis. At this point the project enters the exploratory phase. Weeks of statistical tests and disaggregation on every demographic variable. Folks play with the dataset, graphing every question, maybe in testing out multiple chart types, on the lookout for the interesting patterns and informative results.
Then you get into the communication phase, where you prep your results for discussion. Except the committee wanted the meeting two weeks ago. What’s taking so long? So you slap together the graphs you’ve been generating during exploration and rush into the meeting. Where discussion is scattered, heading down rabbit holes, and unproductive. Everyone leaves frustrated because this whole endeavor was time-consuming, expensive, and ultimately inconclusive.
I’ve been here. Sometimes the meeting turns out a lot better. But it seemed we always had that slap-it-together-this-was-due-yesterday rush at the end. Which compromised the quality of our data storytelling, ultimately under-informing the answers we set out to find.
It happened so routinely, you’d think we would have predicted it. But each time it was like “How did we end up out of time again?”
It’s because we think visualizing only happens after we have the data.
The Revised Process
Good data visualization takes time. So does everything else leading up to it.
The trick is to spread out the visualization process so it doesn’t all occur after data analysis is complete. This might seem counter-intuitive – visualize BEFORE you have the data?? – but if you think about your whole process, you have a pretty good clue about what the data will be before it lands.
Back when you first thought through the committee-generated goals of the project, you were articulating the 3 (plus or minus) big questions the project needed to answer. And you know for certain people will want to see those 3 big answers, disaggregated. You can already start fashioning a slideshow focused on answering those top 3 right up front.
While the committee meets (and meets), you can learn some efficiency tricks and develop templates for your data viz.
As soon as you have the survey instrument constructed, you know exactly what the response options will be that you’ll need to graph. Granted, you don’t have the data, but you can easily get dummy data or data from the last time you ran this survey – right?
While you’re in that quiet data collection period, you can do more than send nudging emails and cross your fingers for a high response rate. You can be graphing.
(Matter of fact, I’ve seen some cases where analysts use dummy data to create some possible outcome scenarios and run those by committee members to get them thinking about how they’d respond. Prepping them for possible decisions and setting them up to take actions.)
That way, when the data comes in and you’ve got it scrubbed up, you just have to replace the new numbers in the graph templates you’ve already made.
You don’t need to wait until crunch time to start graphing. In fact, if you disperse the work of visualizing, you can kinda eliminate crunch time altogether.
When to viz? Always Be Vizzing.