Top Visualization Traps (to avoid)

“The truth is at times we’re all seduced by the idea of immortality through our work, which leads to us prioritizing “newness” over familiarity.”

There are a lot of good reasons the best and brightest of us go off on tangents — the thrill of exploring new frontiers, thinking truly radical thoughts, following a gut instinct, and a host of other very defensible excuses.

But, as data visualization matures and becomes a truly mainstream form of communication for the masses, those tangents often turn out to be “traps” that our minds ignore for the aforementioned reasons. I do them. You do them. We all do them on occasion.

The heart of the problem seems to be that we tend to focus on the wrong things in order to make our work stand out. That’s not always a conscious choice. It happens in subtle ways as we work toward true breakthroughs, innovating or just hitting our deadlines.

When looking at these traps we fall into, you’ll probably find they also apply to many other design scenarios outside the field. I still catch myself stepping right into these holes from time to time.

Here are the top Visualization Traps you should try hard to avoid with data or info

Designing for…

1. Initial impressions

instead of deep understanding.

“You never get a second chance to make a first impression.” That phrase is so prevalent and sage-sounding as we start our professional careers that it’s hard to to ever question it. Perhaps if more people did, they’d realize the other ubiquitous pearl of wisdom “Beauty is only skin deep” is the more appropriate one to heed when it comes to conceptualizing how to communicate the true value/message contained within non-trivial data sets.

A quick spin through many current examples of visualization and infographic pieces begins to feel like they were (perhaps unintentionally) designed to focus almost solely on initial impact and attraction vs. surfacing key insights or significant detail where appropriate. When designing the experience of consuming and sharing our work, it becomes apparent that it’s often that second and third wave of thought and discovery that truly excites people once the initial splash fades.

Yes, it is much harder to sell based on structural integrity vs. curb appeal. Not saying the work shouldn’t be stunning on first inspection — it needs to be to compete for attention and be noticed in this world of ours. That said, it can be stunningly simple, stunningly clear, stunningly moving. There needs to be that critical balance of aesthetic and approach + surfacing insights to jump start anyone’s deeper understanding of your point.

To our great advantage, the medium we work in is capable of so much more than we sometimes use it for. Great design, interactivity, animation, sleek packaging, and clear data representation techniques are our tools. Let’s use them to balance, not bowl over.

2. Novelty

instead of well understood patterns.

If we come up with something truly original, they’ll all sit up and take notice, right? Exactly. So, let’s do this thing unlike anything else that’s ever been published — people will love it. I already love it. And that’s all that matters in the end really. Am I right?

Facetiousness aside, there are countless examples of people delivering work that is different just to be different — rather than embrace something familiar to the audience which requires no decoding or prolonged period of starring blankly at the imagery.

Don’t get me wrong — I know as well as you true breakthroughs sometimes take a strange path. Sometimes it’s a novel approach that pushes us all forward in positive ways. Other times it’s that experiment that failed which helped us learn more than if we had succeeded.

Just be honest about what your goals are.

Don’t confuse novelty with innovation. If you’re not careful, you’ll fall into this trap filled with very pointy feedback from very puzzled people. Wouldn’t a simple bar chart have worked there? ouch.

Your customers and audience will value, love and respect your work if they can understand it without friction. Trust me.

3. Cleverness

instead of predictability.

I knew you were going to say predictability is boring. It kinda is, I know. But, guess what — you’re smarter than alot of people consuming your work. Let’s come back to that point in just a sec.

What is the difference between cleverness and novelty anyway? Aren’t they the same things? Nope. Not even close in this setting. Being clever in visualization and interaction provides something unexpected but very satisfying — often in a “wink wink” insider joke kinda way. Being novel on the other hand really refers more to the new tack and approach taken. It’s newness is the thing. You might hate it. But it’s novel alright.

Now, back to that thought about how you are more clever than alot of other people looking at your work. You know exactly what that monstrously large and complicated data set is telling you. So, why not project that exceedingly clever insight of yours onto the final deliverable?

Well, for starters even though many people will get it and appreciate your command of the material, others will have been much better off consuming something they find familiar, predictable and well understood. Just like with novelty, they’ll spend more time decoding your clever take on the data set or infographic rather than truly absorbing what you’d like them to walk away with.

Predictability is not the enemy of coolness.
Food for thought.

4. Data density

instead of surfacing insights.

“You’re not really going to leave all that blank space, are you?” is typically how the conversation starts.

Data Lovers crave high density layouts to soak in all the facts they can handle. Designers on the other hand love their whitespace, negative space, airy layouts to let things breath, and just to piss you off — using large beautiful typography. It’s an easy argument for the Data Lover to win, though. How can you argue with the logic of having more high value data on the screen or page rather than wasting it on a pretty layout? I mean really. It’s no contest — until you consider we have been going about the display and visualization of data completely backwards for decades.

Surfacing insights (that are not readily discernible by the causal observer) is soooooo much more important these days than jamming as much data into a visualization as possible, I almost don’t know where to start. People are busy. They are not domain experts in everything you are. They are not as familiar with the subject matter, data, patterns, anomalies, and outliers as you are.

So, please — JUST TELL THEM what you gleaned by focusing on summarizing the key points in visual insights. You don’t have to remove the dense stacks of data completely, but at least don’t lead with them. Use progressive disclosure to aid in the discovery and internalization.

What’s that you say? Dense insights. Hmmm…

5. Parameterization

instead of flexible exploration.

It’s relatively easy to parameterize an interactive visualization so that certain properties are configurable at consumption time — sliders or knobs of some sort appear to allow easy value changes within the prescribed range. In reality though, it restricts as much as it helps.

Falling in to the trap of allowing only pre-determined pivots is paramount to leading the witness. Don’t give us predefined sliders that keep me on rails. Give me the ability to be truly curious and explore your data more fluidly. Yes, I know that’s really hard. It’s why we pay you so much ;-)

Try to consider more carefully how you’d allow for more organic explorations of the curious. Don’t just pick several properties to parameterize and call it good. Look for ways to open up the possibility of exploring all the different dimensions of the piece. Sounds like a ton of work, I know. But, it’s so worth it when you see people’s faces light up.

Yes, it takes infinitely more time than just isolating the key dimensions and slapping some sliders or drag actions in there. But, you’ll be setting the stage for the next big leap in visualization — freestyle exploration.

6. Mouse-centricity

instead of multimodal input.

Most of the visualization pieces out there today rely upon either Mouse/Keyboard (or in the enlightened cases, Touch-first) interaction, but few are adept at allowing or even encouraging multimodal input.

We’re at that point in history where people believe they can poke, swipe, talk to, or even command things on their computing devices to do what they want. It’s only natural to want to interact with something in that way. Unfortunately, the truth is that voice, gesture, pen and gaze have not been deeply integrated into our playback platforms enough to encourage authors to leverage them.

But, the wrong thing to do here is to throw up our hands and say we’ll wait it out. The trap is falling into that “wait until its a standard” mentality. Your audience already expects this stuff to work just like all their other digital stuff. And furthermore, they would never articulate this way, but they do expect more than one input model to be possible at a time.

We are a multi-tasking, multimodal society. Our pieces should reflect that.

7. Competition

instead of true collaboration.

There’s nothing wrong with competition. Every industry and market segment has its share of healthy competition. It pushes us all forward, forces us step up and bring our “A” game to truly compete with the best every day. Although, it is a bit different in high tech (and certainly in the data viz space). We share code, best practices and are quick highlight great (and bad) examples.

We also reinvent the same things all the time.

In our effort to be noticed, novel, and clever, it seems we spend less time doing mutually beneficial collaboration with our frenemies and more time trying to design and build a better mousetrap for the fourth time.

There’s no question a real need exists to differentiate our work (whether to land that next client, be published or deliver that killer preso). That said, it’s been shown time and time again that we are always better together. Few things we do completely on our own is truly better than what can be accomplished with the creative spark and push of our friends, competitors and audience/customers working together.

Still waiting to see the mashups, mixups, and true breakthrough collaboration in visualization we see in many other fields.

8. Beauty

instead of clarity.

With so many talented Designers and Data Scientists on the scene there’s been a real emphasis on producing high quality visual appearance of new infographics and visualizations. That’s incredibly great of course, until that focus on aesthetic takes away from effectively communicating. And therein lies the dilemma — we need BOTH beauty and clarity. The trap is continuing to choose visual appeal aspects over nailing the communication of core concepts.

“Clearly Beautiful” viz needs to become “Beautifully Clear”.

And let’s be honest, clarity is not that easy to achieve quickly in people’s eyes and mind. Reaching that critical “moment of clarity” sometimes requires a bit of discovery and disclosure. Think about those time lapse videos of a beautiful flower opening you’ve seen on TV. Flowers are naturally gorgeous to look at while closed and not moving. Yet, when you see them go through the motion and process of blooming right in front of your eyes it’s absolutely breathtaking. That feeling is what we should strive for.

The beauty we seek is not found in first impressions, but rather in how the truth reveals itself.

What are the visualization traps you’ve fallen into?

ABOUT

Mike Pell is leading Design for The Microsoft Garage, an innovation accelerator turning employee’s wild ideas into reality, worldwide.
Bold, insightful and uncompromising, Mike is recognized as a thought leader in the field Visualization and Smart Information.

Learn more at futuristic.com
@mryankeeslicker