Some researchers from Northeastern University have been tracking the whereabouts of some 100,000 mobile users and found some interesting results. Many stories about the research have focused on the privacy implications, but we’ll ignore them for the time being, and focus instead on the main finding: that most people in the study (from what’s identified only as an industrialized nation somewhere in Europe) don’t move around much, and when they do, they tend to go to the same places.
The study found that nearly 75 percent of people stayed within a 20-mile circle for half the year, with the vast majority keeping to an even smaller one most of the time. Then, when they do travel — near or far — they tend to go to the same places again and again. There are potential implications for this data well beyond our industry, but what it’s mean for mobile services?
Russell’s made this basic point before, that most of us tend to spend most of our time in places we know best, and this study underlines that. So does this mean local search is a little misguided, that maybe we don’t need help finding places near us as search vendors would have us believe? I’ve always been a little bit skeptical of the local search market, mainly for this reason. That said, I think there’s still a lot of room in the market for services and applications that help us interact with our local area better. Think things like Socialight, Loopt, Buzzd, or Whrrl.
Conversely, when we do wander out of our usual haunts is when we most need the sort of help that local search or other LBS can provide. So what can be done to better recognize, or even predict, when users need some assistance or guidance? And how should the user experience change for LBS or local search when your users aren’t familiar with the area?
Interesting stuff to consider, I think. This is a space that hasn’t yet been cracked, and it’s still early days. But this sort of research can provide some particularly useful guidance to LBS developers.
Excellent blog. And you’re right on the money with regard to local LBS services. If people don’t need local search (or even if they do) what’s the revenue model behind the current players mentioned above? (Loopt is a subscription play) And even if they can figure one out - is it really going to be sustainable?
Time will tell.
Very interesting article. This study is consistent with users behaviour on MobiLuck. A large majority of MobiLuckers prefer to update their location manually (privacy) by clicking on a place in their list of favourite places (only a few places, home, office, school, friends, restaurants, …).
[...] Local Services For Local People [via Zemanta] [...]
LBS fails because it is not used. Obvious thing to say I know, however it is not used because “most people know where they are most of the time”, as per the research. Thus when they are _not_ somewhere familiar they do not have the learned behaviour to ask for help from their mobile phone. they are likely to ask someone or go online either just before they travel, or whilst there.
thus for LBS to succeed they need to find ways to help people remember that they can ask their device where they are …
and this is where local services may help, if you have a local search service that helps you out in your day to day life, you start to learn the behaviour of asking your phone for help.
Steve
Hi, could you point us to the research paper?
ah silly me…i see the link now. thanks!
Even though people might be in few places, services are national or international and to provide e.g. local weather and news information it’s still very handy to know where the user is (at least with cell precision).
Clearly, a weather information service won’t cover just Hoboken (that would hardly be profitable), but all of USA, etc, so the phone or the network needs to tell the service what info the user is likely to be interested in.
In Sweden we have one and soon two providers of operator-independent nation-covering cell positioning services. A lot of value services can be piggybacked on that.
Generic local search is overrated. Rather think local information.