Tag Archive for user interface

Mapping Faux Pas #5: Bad Data Visualization

This week brings us to #5 in our Real Estate Mapping Faux Pas series: Bad Data Visualization. This is one of my favorites and a mistake I see so many people committing without even knowing it. As a disclaimer, I will admit I’m a bit of a data visualization zealot so bear with me, but I can’t emphasize enough how far good data display goes towards enhancing the usability and comprehension of information, particularly with complex data sets. That being said, let’s talk mapping data visualization, why it’s important, and some of the ways people get it so wrong.

First and foremost, maps are an extremely intuitive way for people to access and analyze data, particularly large data sets and/or multiple data sets simultaneously. As humans, our brains are hard-wired to understand maps, and location is a universally understood characteristic. So if maps are theoretically so great for working with information, why do they have such a bad rap for being hard to use? In short, Bad Data Visualization Techniques! Poor data display tactics can render even the best conceived mapping application useless; it’s time we stop blaming the map for its usability shortcomings, and start blaming the people making the maps.

Avoid Information Overload Avoid Pie Charts on Maps
Information Overload on Map Avoid Pie Charts on Maps

Now that we’ve isolated the root cause for many of the confusing maps out there, what are some of the most common mistakes? The first, seen above and to the left, is putting too much data and too little legend information in one map view. This leads to information overload and makes it virtually impossible for your users to make sense of any of the data sets displayed on the map. The well-known real estate website pictured is attempting to show population density with 11 classes of data, school locations, featured properties, and properties for rent. Not only is there way too much data, but it’s not clear what each symbol / color represents. And 11 classes of data? That’s way too many for the average consumer website.

While the map on the left is clearly difficult to understand, and ugly, the map on the right may initially seem appealing, however it is nearly as bad as the map on the left. First, in the data visualization world, pie charts are a big no-no, so putting pie charts on maps, is a bigger no-no. The pie charts obscure one another and it’s virtually impossible to distinguish what the colors and size of the pie chart represent, not to mention what the actual values being measured are. This is a perfect example of just because something looks cool, doesn’t mean it’s the optimal means to display data. In fact, often the best map visualizations, and the most difficult ones to create, are the simple ones.

To keep you from creating “mapping don’ts” due to bad data visualization, I’ve summarized some helpful tips to keep in mind as you’re organizing, styling, and displaying data:

  • We’ve mentioned this before, but it applies here as well – before you throw data up on a map start with the end in mind. Think about who will be using this data, and what conclusions you want to help them draw from it.
  • Before everything else, make sure your data is clean and you understand its nuances – does your data set have outliers? How is it distributed? Where does your data have holes? Exploring and getting to know your data will help you select your display scheme.
  • When classifying your data, or summarizing it, use a limited number of classes and intuitive color schemes.
  • Resist the urge to show all your data on the map at once – in lieu of 30,000 points on a map, summarize your data to easily understood geographies to make it less overwhelming for users. When you select your geographies, select ones that mean something to your users.
  • Give users creative options to explore your map data such as animations, drill downs, etc.

The most important takeaway is that even if you have the slickest, fastest, easiest to use mapping application, and have avoided all our other Faux Pas up to this point, you must invest the time and thought into proper data display to prevent your site from becoming a major Mapping Don’t! Now that you understand some of the most important tips for good map display, you too can banish ugly maps and put good data visualization principles to work for you to enhance your user experience and make your real estate site stand out from the pack. And be sure to check back next week for a discussion with Skip on how to take your map beyond just the property location.

Mapping Faux Pas #3: Dumb Maps

Now that the real estate mapping genie’s out of the bottle we’re facing some pretty hefty spatial expectations from consumers. In our first few Faux Pas, we touched on the relatively basic concepts of having a map and making it easy to find, and then making sure you’re using an accurate location for the data displayed on your maps. This week’s Faux Pas elevates our mapping savvy and addresses how to avoid Dumb Maps.

What are dumb maps and why are they so bad for your site? Simply put, today’s users expect their maps to be more than just pretty pictures. When you include a map but it doesn’t do anything, it’s a tease at best, and at worst is a reason for visitors to leave your site. Consumers expect the maps they encounter online to be interactive. Now that they’re familiar with maps, they want to use them to explore an area, drill into deeper data, and dynamically answer their real estate questions. Maps that don’t allow users to interact with them are dead-ends, and will ultimately impede your efforts to fully engage your site visitors.

So what’s the secret to avoiding dead-end maps? It’s really fairly easy. If you go to the trouble to plot something on a map, make sure you include features that let users get more information about it. Some of the most basic ways to do this are with standard mouse-overs, on-click balloons, and links to detail pages and additional data and features. Taking this approach will keep you from overwhelming your users with too much data, and yet will quench their thirst for additional details when they find something of interest. You can see an example of this below on the left. Shown is a residential home price trend map. The data is nicely summarized to neighborhoods so you can quickly tell which neighborhoods have had large price drops (red) and which have held their prices (green). When a user clicks on a neighborhood, they can see additional stats for that neighborhood and even access related data sets.

Home Price Trend Map Map with Results List
Home Price Trend Map Map with Results List

Similarly, just as you want to incorporate interactivity to your maps, you also want to tie your maps to the more traditional world of data display, otherwise known as the tabular world. Above, on the right we show a map with a results list beneath it. It’s important to link these two features so that when a user clicks or hovers over a house on the map, it highlights in the list below and vice versa. Taking this a step further, we love to tie our map data to graphs so we can easily switch among data visualization methods and quickly explore our data.

Below are a few tips and techniques for including interactivity in your maps that will delight even the most demanding of consumers:

  • If you have a map with data, even a point, let users get more information by clicking on it
  • Give users map-search options using selectable areas, custom drawn boundaries and to map extent check boxes
  • Bridge the spatial and tabular worlds by making sure your maps are interactive with your search results lists and charts and graphs
  • Let users drill into maps for additional views and layers of information
  • Achieve mapping guru-status when you embed animated maps to show trends over time!

Remember, avoid dead-end maps and the inevitable user disappointment that ensues by building in interactivity to your maps and linking your map experience with the non-map data world. And with that mapping faux pas safely checked off your list, you’re ready to address our fourth common mapping mistake: “Loading Map Syndrome.”  Check back next week for a spirited discussion with Skip around online mapping performance and how it can go so wrong!