Bringing Emotion to Data

Data: Memorable or not?

When I’m not act­ing as a user expe­ri­ence designer and con­sul­tant, I archi­tect and over­haul pre­sen­ta­tions and have done so for almost 2 decades. For about 10 of those, I worked at a retire­ment plan ser­vices firm where, as you can imag­ine, pre­sen­ta­tions con­tained quite a bit of data.

As I’m sure you can also imag­ine, the data was pre­sented in a less than excit­ing man­ner. The emo­tions that they cre­ated were often bore­dom cou­pled with a strong desire to check your phone and get lost in your Face­book feed.

Many peo­ple who present data on a reg­u­lar basis are so famil­iar with it that they assume the data speaks for itself, and so they don’t DO much with it. Guess what? Usu­ally, the data doesn’t speak for itself.

Missed oppor­tu­ni­ties abound in these kinds of decks. How did the data relate to each other? To the audi­ence? Why does the audi­ence care about XYZ when you’re point is ABC? It’s so easy to fall into the trap of clean­ing up data slides so that they’re JUST read­able or on brand and for­get that, as speak­ers and pre­sen­ta­tion pro­fes­sion­als, we should be look­ing for the story that’s hid­ing in the data. We should try to attain a deep under­stand­ing of the charts and tables in front of us so that we can present it in a more impact­ful way.

What makes data memorable?

It tells a story. It’s inter­ac­tive. It makes you want to dive deeper.

These are things that make data more mem­o­rable and hope­fully con­tribute to your audi­ence act­ing out the desired call to action. But when we start talk­ing about data specif­i­cally in the pre­sen­ta­tion set­ting, there are a lot of fac­tors work­ing against us before the pre­sen­ter even walks to the front of the room.

You’re already lim­ited by time, of course. But other fac­tors might include things like the mind­set of the attendees.

• Do they want to be there? Or do they HAVE to be there?
• Is it around lunchtime and they’re either starv­ing or in the mid­dle of a food coma?
• Are the atten­dees famil­iar with the topic or is this all new to them? 

Memorable data in presentations

Con­sid­er­ing all of these fac­tors, let’s revisit what makes data mem­o­rable and ask, “What makes data in pre­sen­ta­tions memorable?”

For data to be mem­o­rable in this sce­nario, you need to be strate­gic. First, is the infor­ma­tion under­stand­able at a glance? Slides are like bill­boards; the data needs to be deliv­ered as though it’s on one.

Sec­ond, is your data read­able? How big is your room? Will the peo­ple in the back be able to read it? How small is the screen peo­ple are watch­ing your pre­sen­ta­tion on? Test your pre­sen­ta­tion for read­abil­ity in a big rooms and on small screens. Pay atten­tion to font size, color con­trasts, and line thicknesses.

Third, is the data relat­able? Does the data evoke an emo­tion or is it B‑O-R-I-N‑G? If you play your cards right, you can make data impact­ful and mem­o­rable with hardly any “design” at all. Some­times, all you need is the right tim­ing in your deliv­ery to bring atten­tion to an aston­ish­ing “what’s in it for me” sta­tis­tic or line chart.

Recognizing data clues

Before we get into how to design a data slide to evoke emo­tion or inspire action, let’s first talk about how to rec­og­nize a data slide when we see one. Of course, there are very obvi­ous exam­ples like tables full of num­bers, bar charts, line charts, and pie charts.

Some exam­ples aren’t quite as obvi­ous. Tables that are full of text, tables depict­ing types of nouns, and slides full of bul­leted text. Sure, maybe this one says 15% at the bot­tom, and there are a few num­bers in here… but what’s actu­ally going on?

As you read through the slides you’re given, watch for data clues. Words that indi­cate that there is a pos­si­bil­ity that the infor­ma­tion on the slide can be rep­re­sented in a more visual and quan­ti­ta­tive man­ner. After we iden­tify these slides, we can start think­ing about how to design for an emo­tional response.

Data clue words:

• Per­cent
• X in X did Y
• Num­ber words
• Frac­tion words
• Time words
• More than and less than rela­tion­ship phrases
• Trend
• Phase
• Geo­graphic location

This is an actual slide I received from a client with a list of date ranges. But the way this is writ­ten, can you look at this slide and feel how long these time peri­ods are?

Visu­ally, we can cre­ate a time­line that ulti­mately shows how each of these phases relate to each other. How fast will the Con­fer­ence room fur­ni­ture deliv­ery period feel com­pared to the AV install? Let’s find out.

Once we have this info on a time­line, we can see how long these phases last with­out hav­ing to think about it. Our first feel­ing we bring to this data is clar­ity. Depend­ing on what depart­ment your audi­ence mem­bers sit in, they may also get rather excited if they see that their part of the move is about to start.

Tak­ing a step back, we can see where we came from and where we are now. By refram­ing how we present the infor­ma­tion, we can already improve the chance that our audi­ence feels the emo­tion that is highly sought after– clarity.

Emotion in data

Now, when I talk about bring­ing emo­tion to data, I’m not nec­es­sar­ily talk­ing about any­thing extreme. Sure, it would be awe­some if we had some­thing to present that was so pro­found that, through our design and deliv­ery, we make the whole room break down into sobs of joy, or rages of anger that fuels a momen­tous call to action. But in most cases, we’re aim­ing to cre­ate lit­tle tiny sparks of one of these (slide with emo­jis). Well-timed hits of emo­tion pro­duce a higher like­li­hood of infor­ma­tion reten­tion and a greater chance of audi­ence mem­bers tak­ing the desired action. What we don’t want to do is make them feel over­whelmed, bored or sleepy.

There are a few design tech­niques that you can apply in var­i­ous com­bi­na­tions to achieve the desired emo­tional reac­tion includ­ing color, imagery, lan­guage choice, and motion. You don’t need to apply all 4, either. If one or two tech­niques are applied really well, the data take­away will stick with your audi­ence a lot longer than if you just put a chart on a slide.

Color

Color can poten­tially give you the most bang for your buck. You don’t have to worry about copy­right or put as much thought into is as you would for reword­ing titles and such. How­ever, if you have super-strict brand guide­lines when it comes to color, some­times hit­ting the mark here can be tricky and you have to get cre­ative with color substitutions.

Imagery

Images can really bring a lot of impact to your slides. Then can con­vey a plethora of emotion.

Each of these images makes you feel some­thing dif­fer­ent and at vary­ing lev­els of emo­tion as well. There are some images here that can be inter­preted in more than one way. Does the path rep­re­sent the begin­ning of an excit­ing new jour­ney or does make you feel lonely and iso­lated? Depend­ing on the data each image is paired with, it could be per­ceived as either one.

The dessert conundrum

To fur­ther illus­trate how to design around ambi­gu­ity let’s use the fol­low­ing sta­tis­tic. The sub­ject of this data point is con­tro­ver­sial, so pre­pare yourselves.

“Clowns eat cake 65% slower than kids.”

This is an inter­est­ing bit of infor­ma­tion. Ok, awe­some. So, we’re talk­ing about clowns. Let’s start design­ing this data to cre­ate impact. First, let’s use imagery for this since there are plenty of clown pho­tos read­ily avail­able. And we’re talk­ing about cake — let’s rep­re­sent that per­cent­age with a group of icons.

There! We have a happy delight­ful slide about cake and clowns! I feel like the pre­sen­ter will be com­pletely happy with this data slide. It’ll make the audi­ence hun­gry for more. Or at least for cake. Unfor­tu­nately, because my pre­sen­ter worded the sta­tis­tic in an ambigu­ous man­ner, it appears that I designed this exactly wrong.

What the pre­sen­ter REALLY meant is that clowns eat cake slower than the EAT kids. Data can be ambigu­ous. Because of that, we want to make sure we’re sup­port­ing that data with some visu­als that help the audi­ence truly under­stand the hor­rific real­ity about these par­tic­u­lar clowns.

We can see now how we’ve made a pretty big dif­fer­ence at how our data is inter­preted and felt by our audi­ence just by using color and visual assets. This all brings us to the one thing that would have helped me avoid this problem.

Language choice.

Besides ambi­gu­ity, there are other things to take into con­sid­er­a­tion with regards to lan­guage choice. Is your lan­guage too strong? Make the data story com­pelling with­out yelling. Know the data and its pur­pose. Empathize with the sub­ject mat­ter and audience.

If we tweak the lan­guage used in our clown exam­ple, we can make a much stronger impact. The mes­sage is no longer ambiguous.

Motion

Motion is very pow­er­ful. Never use motion lightly. If you have some­thing that is the main call to action or a piece of data that is a really big deal, that’s where you use motion in a very inten­tional man­ner. Motion can be implied like this map by Charles Joseph Minard. Prob­a­bly the best sta­tis­ti­cal graphic ever drawn, it por­trays the losses suf­fered by Napoleon’s army in the Russ­ian cam­paign of 1812. Begin­ning at the Pol­ish-Russ­ian bor­der, the thick band shows the size of the army at each posi­tion. The path of Napoleon’s retreat from Moscow in the bit­terly cold win­ter is depicted by the dark lower band, which is tied to tem­per­a­ture and time scales.

Hans Rosling was a mas­ter at using motion to help peo­ple under­stand com­plex data sets. Let’s watch. I encour­age you to watch the whole video, but I tend to focus on the snip­pet between 00:30 and 3:20. What keeps you engaged?

You can design for emo­tional impact all you want. You can do every­thing I sug­gest to you today. But if the pre­sen­ta­tion is deliv­ered poorly, the infor­ma­tion can still fall flat. Deliv­ery is key.

For before and after exam­ples of data makeovers, watch the first video above.

Take care of your data

Main­tain data integrity. Never let design get in the way of data accu­racy. Don’t use stacked icons as bars in charts. Don’t use 3D charts. Even. This is an extreme exam­ple of the same data side-by-side in 3D and 2D used to illus­trate a point. How­ever, even a 3D pie chart dis­torts the rela­tion­ships between the sizes of the slices mak­ing it dif­fi­cult to deter­mine how to read the data.

Beware false nar­ra­tives. With great power comes great respon­si­bil­ity. Some­times we see a story where there isn’t one. As a data sto­ry­teller, you have a lot of poten­tial power and respon­si­bil­ity. How you select, arrange and present the facts can change the way peo­ple think or how they view the world. Cog­ni­tive biases can twist data stories.

Truthi­ness: the belief that a par­tic­u­lar state­ment is true based on the intu­ition of an indi­vid­ual or group, with­out regard to evi­dence, logic, intel­lec­tual exam­i­na­tion, or facts. Truthi­ness is bad. Accu­rate rep­re­sen­ta­tion of data in a non-lead­ing man­ner is of utmost importance.

Cor­re­la­tion does not imply causation.

Just because data sets look sim­i­lar doesn’t nec­es­sar­ily mean they have any rela­tion­ship what­so­ever. Here’s a great place to go for more ridicu­lous charts.

Go forth and present data well.

There you have it. I hope this has helped you improve how you present data. Remem­ber, by using color, imagery, iconog­ra­phy, care­fully cho­sen and clear lan­guage, motion, and deliv­ery you can get your audi­ence excited about your data.