Why a map isn’t always your best choice for a data viz | by Information for Humans! | Sep, 2021


Make sure you also present the things that matter.

Photo by Martin Sanchez on Unsplash

In 1957, French film director Albert Lamorisse invented the board game Risk. Gameplay takes place on a map of the world, where players move markers to track the territories, continents, and sea lanes they control. The goal: world domination. All the world, right before you, right in your sweaty palms.

When it comes to data visualization, are you playing world domination? There are too many data visualizations that use maps, losing our audience in irrelevant, distracting information. Most of the map visualizations I see could be better shown in alternative, simpler, more informative layouts like tables or graphs. Let’s take a look.

Maps help us visualize location, size, and distance between things. We’ve been using them since the beginning of time to find our way around the world.

Militaries use them to show landscape and adjacency — one of the most important aspects of the battlefield strategy. Where might I enter a battlefield with the least amount of effort? Or where, alternatively, might I obtain an advantage by placing my forces in an elevated terrain?

Israel’s location relative to its neighboring countries is a perfect example. Compare Israel’s position on a basic cartographic map against the same situation, expressed more clearly by in the landscape of the region:

Source: Wikipedia Commons

Unless an essential part of your job or mission involves navigation and placement of resources in natural or urban terrain, let’s consider what we need for an attention-getting, action-demanding presentation. Data visualizations help you quickly highlight and direct your reader’s attention to the most important, immediate issues that require their attention:

  • A poor performance relative to goals;
  • A risk of financial, operational, or experiential problems;
  • An opportunity to achieve or surpass a goal.

People use maps all of the time in data visualizations. For example, sales and profit, by territory or region. Tableau, for example, offers up this chart on its own how-to pages:

Source: Tableau Public

It’s eye candy, but what does it actually tell us? It pretty much just shows us where big, profitable markets are. But — does geography specifically influence that? Looking at the three states adjacent to California, I’d guess: no. All I’ve learned is that California is big scale, big margins.

Considering my three basic reasons for a data visualization, does using a map for sales and profit tell me:

  • Performance relative to goals? No. Even if the author added information about whether California is above, below, or at my target sales and profit, the format of the map, and California’s location on the map wouldn’t give me any further insights.
  • Risk of financial, operational, or experiential problems? Nope. Same thing. Even if the author added information about where wildfires, earthquakes, and mudslides are more likely to occur, these are unusual events, and not ongoing factors in our month-over-month sales operations.
  • An opportunity to achieve or surpass a goal. We might use maps at a much higher resolution, say in zip codes, to compare adjacent sales of nearby stores, and make some conclusions about whether we are over-serving or under-serving markets, based on consumers’ traffic patterns or daily habits. But at this level of resolution, showing all fifty states? Nope — doesn’t give us any relevant or pertinent information.

Lots of companies want to show their COVID-19 return-to-work strategy. Take a look at the map, below. It provides us with four different visual elements, each communicating important information:

  • Location of the company’s facilities;
  • Size of the facility (in seats);
  • Status of COVID-19 in that state;
  • Status of the facility’s readiness to deal with the pandemic.
Source: Author

Seeing this all on one page, on a map of our entire domain..whew! It feels like the presenter has control of the situation! Here, our attention is drawn to the red, large elements in the map. Arizona looks problematic. Missouri and North Carolina, too. But wait — those states are experiencing lower four-week mortality. Are we really clear what we need to do first? What actions do we need to take?

Sure, knowing the COVID-19 situation in the state where our facilities are located is important. But is showing the map, where we see the locations of the facilities relative to each other, really that important? Or are we just creating the illusion of domination over the situation?

What if we added some more relevant and pertinent information about the situation in a different format?

Source: Author

Here, we learn a few things we couldn’t see easily on the map:

  • We actually have 20 locations to worry about;
  • 15 of them are in states with high or medium COVID-19 mortality;
  • We added a time dimension — and this shows that five of them are not ready to be opened — representing 27% of our 10,000 seats;
  • It’s July now — three of them are not ready to be opened for another two months, and two of them won’t be ready until November!

I now want to know some things about those five that aren’t ready. Where are they? What else do we know about them? Depending on whether the dashboard is interactive, or if we have additional pages of detail, we could get more information. In this case, the dashboard allows us to click down into the five that are not ready, and we get more information:

Source: Author

Ok — so now, we know specifically that we have three that won’t be ready until September, making up 16% of our 10,000 seats, and two that won’t be ready until November, making up another 11% of our 10,000 seats. That production facility in Texas looks rather important — we’d better get on it and make sure we have it under control.

There’s a reason why Risk was such a successful game — people love being the master of the universe! Maps are eye candy for those who want control. And what manager doesn’t want to be in control of the situation? A map will surely get their attention, so go ahead, allow it to make it into the spec.

Source: Author

But make sure you follow these layout and design principles, below:

  • Give them more than a map. Provide them with clear facts on the total scope of the issue, and what matters the most.
  • Consider dimensions that maps can’t easily communicate, like time. Time is one of the most important features of any data visualization. It tells us when something happened or will happen, and whether we still have time to act.
  • Add detail that allows people to get to the heart of the matter. Whether it involves point-and-click descents into detail, or subsequent pages, you’ll want to give your audience the opportunity to choose different journeys into the information, and return with what they need.

With those principles in mind, you’ll indulge everyone’s desire to dominate the world…and maybe you’ll just end up helping people focus on important things, too.



Source link

Latest articles

Related articles

Leave a reply

Please enter your comment!
Please enter your name here