Foursquare experimenting with recommendations engine and “manufactured serendipity”
Each day in my office we face a challenge: where to eat lunch. We’ve tried building elaborate spreadsheets to track the local eateries and where we’ve been recently, but in the end we usually end up eating at the same places over and over again because picking somewhere new is just too difficult.
It turns out foursquare has already built a tool for exactly this purpose that they’re using in their New York office, co-founder Dennis Crowley told a packed house at Picnic in Amsterdam earlier today. It’s part of their experiment with using the vast amount of data they have about where we go and who we go with to build a recommendation engine.
Their internal lunch tool considers the nearby places that are popular at lunch and what’s on their to-do lists and filters out the places they’ve been recently to recommend somewhere for them to eat lunch. Crowley hopes other developers will be able to build off foursquare’s data to enable similar tools for all sorts of recommendations.
Another of Crowley’s long term goals is to use your phone as a sensor that’s constantly aware of all the things around you. It could, for instance, buzz when you walk past a sandwich shop at lunchtime — knowing you haven’t already eaten — because that’s a place your buddies like to eat, or a place you previously marked as a place you’d like to visit but didn’t get a chance. This, of course, would require always-on GPS, which has proven to be a huge battery drain for mobile phones. Crowley dismissed that as just “a big technical hurdle we need to get over.”
Also on the foursquare roadmap is the identification of experts on a topic by their number of checkins in a particular category. Users would be awarded experience points as a sushi expert or skiing expert based on their checkins in those categories, for example. Their recommendations could carry more weight and be used to suggest places other users should visit.
These are the kinds of ideas that could help foursquare move past the simple checkin and beat Facebook in the location game. Foursquare has acquired enormous amounts of data about the places we visit, who we go with and what we like that it’s now a matter of them feeding that data back to us in useful ways. The game mechanics, of course, will always be around as an incentive to keep users checking in, but it’s recommendations that could really set foursquare apart from the other simple checkin services.
Here is the full video of Crowley’s talk:
#picnic10: Dennis Crowley (FourSquare) over Social Cities from MGvandenBroek on Vimeo.
/Photo by Renato Valdes Olmos (@Renn) on Mobypicture; used with permission

Pingback: Foursquare Experimenting With Recommendation Engine | ECtimes.com
Pingback: Foursquare Experimenting With Recommendation Engine | TechAggregator.com
Pingback: Foursquare Experimenting With Recommendation Engine
Pingback: Foursquare Will Soon Tell You Where To Eat | The Blog Herald
Pingback: SearchCap: The Day In Search, September 23, 2010
Pingback: Foursquare Builds More Bridges to the Real World: Tech News «
Pingback: Xentience (Internal) Blog » Foursquare Experimenting With Recommendation Engine
Pingback: primus inter pares
Pingback: We Are Social’s Monday Mashup #42 / we are social
Pingback: FourSquare Recommendation Engine « Facebook Ecommerce Blog
Pingback: Game Time for Foursquare « @ the intersection
Pingback: Facebook vs. Foursquare: It's Game Time
Pingback: Quora