At MapQuest, we love maps. We’ve been creating millions of them every day for 15 years, helping people navigate among millions of places across the globe. Maps are tried and true, but at their root, they are really just a canvas for data visualization.
In the past couple of years, the mapping space has rapidly shifted from just a need for “maps and directions” to a need for “local search.” Maps are great for roads and terrain, but local search results displayed as pins on a map or just a flat list could be made more useful.
Finding where to go can be a frustrating experience. Often you find a bunch of restaurants – all with 3.8 stars that require you read through hundreds of reviews to find a new place to eat. Or, if you find a “best of” list, it usually compares restaurants at the city level, which isn’t what you want when you’re hungry now and don’t want to drive across town.
A few months ago, our MapQuest Labs team came upon a new way to tackle this problem by presenting this data in a structure paired with geographic context to help people find what they want faster. The idea is pretty simple: Why make you do all the work to figure out which 3.8-star restaurant is best, when we can simply tell you the best, the second best, the third best, and so on?
How can we do this? It’s because we have a lot of data points; billions of data points in fact. These data points collected over 15 years tell us, among other things, where people are, where they are going, what they are looking for and when they are looking for it.
Individually these data points are not that helpful, but in aggregate they are extremely useful. Analyzing this data lets us give you information to quickly and accurately make decisions.
Today, we’re launching MapQuest Vibe, which is your shortcut to becoming a local, anywhere. This early beta cuts through the clutter of ratings, lists and pins on a map, by giving you actual rankings of places based on key criteria within a local neighborhood context.
The “local knowledge” is generated with a new patented algorithm called VibeRank. This blended social-algorithmic formula takes several implicit signals (like searches on MapQuest and cartographic data), creates a baseline ranking and then layers on explicit social signals from the new Vibe pages.
Based on this algorithm, MapQuest Vibe profiles more than 50,000 neighborhoods, 27,000 cities and 50,000 hotspots in the U.S., reaching 98 percent of the population.
The neighborhood pages let you explore everything a neighborhood has to offer, including restaurants, attractions and services, all ranked in a clear order. Neighborhoods are scored according to attributes like popularity, walkability, and edginess, and the aggregate quality of the places in that neighborhood. These rankings and scores are influenced when you vote up places you like and vote down those you don’t.
You can zoom in further to see the hotspots in a neighborhood. We define hotspots as the places where people and businesses congregate in that neighborhood and have a high density of highly rated points of interest.
Or you can zoom out to a city page to get a sense of the neighborhoods that make up a city. It’s a great way to quickly see where you’d want to spend a day or even get a sense of where you might want to buy a house.
All of this is available today at mqVibe.com and will be available in a few weeks as an iPhone app.
A common problem for mapping and local search products is that the ground-truth is hard to keep up with. When you’re dealing with tens of millions of places, it can take a while for new businesses to be added, closed businesses to be removed or to get everything categorized accurately.
mqVibe approaches this in two ways: First is the traditional way of trying to use the best data providers and allowing businesses to contact us directly to keep their data up to date; the second way is through the social feedback loop inherent in the system. The more people who vote and comment, the more accurate the underlying data. And, in the next few months, we’ll have better tools to let users help us fix this on the fly.
Over time, you’ll see VibeRank applied to more and more things on MapQuest. Whether it’s applied to highway exits, fuel-efficient routes, national parks or even used as a reputation indicator on user profiles, you can expect to see clear and simple ranks and scores to help you make a decision.
Additionally, the neighborhood and hotspot pages can turn into a vibrant engagement platform, which brings local news, events, deals and much more into a useful hyper-local context.
The early beta of MapQuest Vibe, VibeRank, and these new neighborhood, city, and hotspot pages are just a small first step in the next chapter of MapQuest’s journey. Please try them out and let us know what you think.
Vijay Bangaru, VP Product