Big Data

Another quick fix for Apple Maps: MapClutterReduce

One of the features of Apple’s new Maps product that I like is that it automatically puts nearby places on the map as you drive. This makes it easy to see what’s coming up without having to do a search.

But you can have too much of a good thing. And in Apple’s case, it frequently puts too many places on the map. A lot of these places are not the sorts of businesses or venues people will just stumble into while they’re driving. They typically have specific intent.

In the example above, you can see that Apple shows an auto body shop and a smog test station. Those aren’t businesses that people casually frequent.

As a general rule, I would look at three factors to determine which places to show:

  • Is this something people might spontaneously try? Restaurants, bars, cafes, and retail generally fit the bill.
  • Would a typical driver have a need to stop here? Gas stations and rest stops fit this requirement.
  • Is this place a landmark that would help the driver figure out where he or she is? If you think about how people give directions to each other, it typically revolves around landmarks. e.g. “Go until you hit Flatiron Building and then make a left,” not “Go 1.3 miles and make a left.”

If a place doesn’t meet any of those three criteria, it’s most likely clutter and can be eliminated from the map unless a user specifically asks for it.

That’s a rough cut of how to solve the problem.

But there’s a better, big data approach that Apple can take. Apple can use data on what places people search for or visit to figure out which places should show up on the map automatically. This is similar to how Google uses click data as a signal in its search results.

(Disclosure: I own Apple stock.)

This is my third piece on how Apple can improve iOS maps. Also see:

http://venturebeat.com/2012/10/03/a-turn-by-turn-guide-for-apple-to-improve-its-navigation-app/

http://venturebeat.com/2012/10/10/a-quick-fix-for-an-ios-map-problem/


VentureBeat is studying mobile marketing automation. Chime in, and we’ll share the data.