New Yorker John walks out of a successful sales pitch with ABC chemicals inc in Pittsburgh. Elated, he feels it is time to celebrate so he needs to find a decent bar for a well earned drink. He taps “bar” into his mobile phone and up pops the nearest bar to his current location – a quiet, slightly dingy hotel bar – not quite what he had in mind but it would do.
At about the same time, in the same city Barry, another New Yorker, does the same, only for Barry the mobile phone makes a great recommendation – a trendy bar with great music and a few friendly people to get chatting to. He enjoys his evening.
Sounds like Barry had a better time. So what’s the difference between John and Barry? The difference is their data shadow. Barry has been checking in to bars he likes regularly over the past year, he has built up a “shadow” of data about places he has been, and bars in particular. The recommendation engine on the phone is able to cross reference the bars that Barry frequents most in New York with other similar users in Pittsburgh. It leverages a “birds of a feather” principle using both his social data (the behavior of his friends) and his timeline (his behaviour) as well as his location to produce a much better recommendation. Amazon uses a similar technique to offer recommendations on what we should read next.
Creating a data shadow clearly can deliver benefits by making our chosen recommendation engines cleverer. Not just in finding bars but also in finding restaurants, choosing schools, identifying the right garage for our car if it breaks down and so on.
The problem with data shadows though, is that they offer little or no benefits at the time. Building them can be seen as rather secretarial, requiring explicit “check-ins” on Foursquare, or clicking the “Like” button on Facebook.
This problem is solved through gamification by companies like Foursquare – it hopes that eventually its customers will value the intrinsic benefit of “a data shadow” but to get there they need extrinsic benefits to motivate them – hence the badges and mayorships that litter the Foursquare user experience. The extrinsic benefits encourage check-ins that the data shadow needs. For expert users, over time, the extrinsic benefits fade away while the intrinsic benefits will become more important – the user is trained to check-in and gamification’s work is done.
So, if you’ve ever wondered why you are still loyally checking in to every Foursquare place, here’s the reason – you’re building a valuable data shadow so future recommendations will be better. Great, now where’s that bar?