Hyperlocal on mobile is more hyped than hyper, says Yahoo!

The idea of “hyperlocal” mobile services has been gaining a lot of popularity recently – but just what does that mean? At the Location Business Summit this week in Amsterdam, Gary Gale, Director of Engineering, Yahoo! Geo Technologies gave a refreshingly amusing presentation that asked just what “hyperlocal” actually means, and whether it’s really ready yet.

“Location” has been one of the drives of mankind since the first days of communication – after all, you need to know where someone is in order to deliver a message. From smoke signals and homing pigeons to SMS, location is a necessary ingredient. So is a location-based application the pinnacle of that drive? After all, since it’s on your phone, your precise location is reachable at all times. And now you can even interact with your location because of “hyperlocal” search and mapping.

But what IS hyperlocal?

Gary proposes that we are in the age of location-based hype at the moment, and that the phrase “hyperlocal” can mean wildly different things depending on who you ask. He boils down the essential “ingredients” of hyper-local as follows:

- it incorporates entities and events located in a defined area, on the “community” scale
- it is intended to be used for and by the people in that community area
- it is normally created by a resident/tourist for that area

So those are the general ingredients. But what else do you need? You need a matching service to provide content for the right area. As such, you also need content. You need intent to use that content it in a local fashion. You need to have users. You need geo-location for tagging and parsing of content. You need geo-intent (for example, people tend to walk in UK vs. drive in US – so your service needs to take that into account when providing directions). You need the location of the user – which is usually one part Internet Protocol, one part GPS, and one part Assisted GPS.

Put all that together, you start getting “hyperlocal services”. But what kind of categories do these services tend to fall under. There’s the “classic” flavour, with local information being provided through local media. There’s the “corporate” model, like Tesco providing you with directions to the closest retail location. And there’s the “social” type, like foursquare.

Crucially, each of these categories treats “hyperlocal” with completely different levels of granularity. Some take “local” as meaning an entire county. For others, local means a city. Or a neighbourhood. Or a city block. So hyper-local itself can mean a lot of things, depending on what your definition of “local” is. But there are other elephants in the room. First is the matching problem – IP location, which is what the majority of LBS uses, is notoriously innacccurate. And if you’re working on the block-level of local, then a service which thinks you’re 200 feet away from where you really are is useless. Second is the fakery problem – you can fake your location. Foursquare in particular has had problems with this, with some users “checking in” to locations from their couch. Finally there’s the problem of what Gary called the “Geo-Babel” – these different services use different location and tracking systems, which means the same place can be referred to in different ways by different services. There can be 20 different geo-entries on a single location, and information can’t be shared between them.

Final word? Hyperlocal is too hyped at the moment. The industry has a LONG way to go before it starts feeling the benefits that the hyperlocal crowd are already selling.