Foursquare's API means it can be much more than “maps with people in them”

By Tom Limongello, Guest Contributor , written on July 8, 2013

From The News Desk

Fred Wilson, at a recent PandoMonthly, called Foursquare "Maps with people in them." Maybe if this were 2009 it would be. But I'm forced to disagree with the iconic venture capitalist. Foursquare is, or rather could be, much more. I view Foursquare as a combination of three core concepts: friend stream, explore and lists, which owe their existence to a powerful and highly developed API, although that has been both a blessing and a curse. This provides a foundation to Fousquare's potential second coming, if the company plays its cards right, and it involves maps, but not necessarily with people in them.

The current Foursquare interface is a compromise where your friend stream comes up three-fourths of the screen to meet a small map. Dive into the map and if you have friends checking in in the past day, you'll see faces, perhaps faded like Marty McFly's siblings in Back to the Future. You'll also encounter icon-organized recommendations. Tapping on them is quite useful, as you can save venues to a list, check in or share the place with friends. But Foursquare, strangely, never turned into a chat app, and messaging is not possible within it.

Foursquare claims 40,000 developers but that is not descriptive of what or why those developers care about its API. The first location graph use case for developers who wanted the API was based on a feature that Foursquare did away with, because it was way too open, which was to show who was checked in to a particular place, most likely a bar. Sonar and later Banjo and most destructively Girls Around Me  pushed the logic such that they were sharing data for users who weren't really using the 3rd party app but just passively giving Foursquare checkins automatically.

That turned out to be a dead end. One developer, Matt Newberg who was looking to create a way to identify users for his location-based casual games platform,, started with Foursqure's API but then had to move away from it in favor of his own service. He commented on how the early Foursquare API worked:

"Foursquare's API used to be too good to be true," he wrote. "For any set of geo coordinates, you could see all the users who were checked in nearby by venue along with their Twitter and Facebook profiles. I built two games on top of this API that were able to visualize all these nearby foursquare users, but the only way my users could interact with them would be through spammy tweets sent via my Twitter bot since Foursquare had no way for apps to send messages, challenges, or application requests to other foursquare users."

In fact, in the past year, we've seen 3rd party dating apps grow up on Facebook connect like Tinder, Hinge, Acquaintable and Coffee Meets Bagel that basically just use Facebook permissions to enable their own algorithms for surfacing users, and they layer on messaging in their own way. With Foursquare, however, its location setting was too accurate for helping people meet each other – or more simply, location was not the key to relevance.

The second use case was for apps like Instagram that actually augmented Foursquare's dataset. Instagram added checkins from places other than bars, restaurants and offices to better lit spaces where you were more likely to take a photo, like parks, beaches and other more weathered venues. Foursquare didn't benefit as much as Instagram and Path did from this relationship but it increases the number of photos associated with Foursquare venues.

After that, Foursquare probably saw, based on its own hackathons and global usage of its API, that there were lots of API-only apps that delivered their value over email. My favorite, Assisted Serendipity, emailed you when female-to-male ratios in bars were favorable. Rather than have these user interface-free apps to live only in the cloud, Foursquare tried to create a platform approach to give developers an ability to build apps that would live within the Foursquare experience, and called it connected apps.

This created a flurry of light, experimental apps, some of which were so light I had no idea what they did, like, GroupMe, #mom (which texted your mom that you had checked in somewhere, like an airport) and Quora, which looked interesting as a Q&A for locations service but it's impossible to find the correct link to set it up.

If a user's checkin history is a privacy nightmare for Foursquare to support in 3rd party apps, and if connecting apps to live in-the-friend stream is potentially not going to grow third party app user acquisition or even activity, then what is the future of the API?  

Messaging makes sense, like you go into a map and tap on a person's face to start a conversation. But if no one is checking in, then that might be a gamble. If a checkin is not  the only interaction on Foursquare, how about better tools for passing venues back and forth in a Snapchat-like interface, where I could comment on the menu or something more interesting than forcing someone into another map provider?

If Foursquare isn't interested in re-starting the UX of the app or helping users plan for meeting up in the future, that would actually make it even more interesting if they were going to open up the API to better interactions for 3rd party interfaces that might be more interested in experimenting in new map-first user behavior. If you look at Waze, although people aren't specifically engaging with each other, they are talking about what's happening on the map, and there's a ton of activity, but it's not for walkers like me.

Foursquare started as the leader in the location graph, blowing away loopt and other previous services like Boost mobile. It still has the best database, and if you travel outside of the US you'll see that in big cities it's more accurate than Google for addresses (if only it had a connection on the iPhone for using Google maps for directions!). Now, a whole ecosystem of location graph experiments have not been leveraging Foursquare's API. But this could change as we start to use maps on watches, in cars, on tablets and elsewhere. Then, at that moment, we will stop worrying and start to love the map.

Whether we have to give up our location via a checkin or not, we are increasingly going to get comfortable with maps as an interface for directions, recommendations, commentary, and planning. And if Foursquare is going to put itself back on the map, so to speak, there needs to be a powerful new set of tools – either within Foursquare or available via its API to developers.

If that happens, Foursquare would then transform itself into a must-use app again.

Image: Flickr