Information processing and information creation are what defines us as a species. When sitting in a local Starbucks, a blogger might write about what it feels like to be in Opelika, Alabama today, for example. On the way to work in Lewistown, Montana an amateur photographer might take a picture of the local grocery store. And all over the US, teenaged clones of Justin.TV will record countless hours of their surroundings.

The net effect of all of this is the increasing availability of fine-grained information about locales. This information is both interesting and valuable. It is sought after by people living in these places and by advertisers who are trying to reach these people. A handful of startups are recognizing the big potential of local information - relevance. In this post we look at different aspects of the hyperlocality, from satellites to local blogs, and ponder how this information will be organized and monetized.

The picture at the top of this post is the view of my house produced by Google Maps. Even a few years back such picture would be unthinkable, but today we take it for granted and don't make much of it. Satellite imagery is an example of top-down hyperlocal content, but most hyperlocal content is bottom-up because it is user-generated. Photos and video are most familiar. Millions of users all over the world are taking snapshots and streaming videos of their surroundings.

At the start of the social media movement people got excited about tagging their content. While this is an effective approach, it is not always precise. Lately we have seen more and more tools that allow users to specify the location of the content more precisely. One of the more popular techniques is geotagging or geocoding, in which users typically specify location using latitude and longitude coordinates. The corresponding microformat, called Geo microformat, allows people to geocode photos which are embedded into web pages.

Admittedly some of the user generated hyperlocal content is of questionable utility. The video above shows a train leaving the Opelika, Alabama station. Besides serving as proof that the place exists and that there is at least one person with a video camera there, there is not much to it. But we would feel differently if the video had instead captured a local crime. This example points to a big challenge with user generated hyperlocal information - filtering and organizing it on a massive scale.

Because there is so much of this kind of information out there, distilling the truly interesting content is a big challenge. Because this information is created by people all over the world using different tools and technologies it requires a combination of mining techniques. User generated tags, titles, descriptions, geo information as well as sophisticated image and video processing tools have to be used together to make sense out of all this content.

Self-Organizing Hyperlocal Information

Perhaps one of the most innovative and interesting usages of hyperlocal information has been developed by a startup called Dash. Launched at DEMOfall 2006, this company is creating one of the first distributed traffic systems for drivers. At the heart of Dash is a simple idea of having a car computer connected to the grid in real time.

Instead of relying of local news to report traffic jams, Dash drivers get traffic updates from each other. Each Dash system is linked into the local network with other Dash drivers, acting as a single node. The traffic information is automatically inferred by centralized software based on the location and speed of individual cars. Clearly, in order for this system to work, there needs to be a sufficient number of drivers with the Dash system installed on the road in your neighborhood. Yet, this example shows a potential power of automatic hyperlocal information.

Hyperlocal Classifieds

Local photos and videos may still be new to many of us, but not classifieds. Hometown newspapers have been in the business of connecting people to local real estate, cars and jobs for quite some time. When Craigslist came online it changed the rules of the classifieds game. And recently Craigslist went mainstream. We looked at job and real estate listings in Montana and found them quite vibrant. In all cities and areas where the site is available local newspapers are losing their business, because in most places you can post to Craigslist for free.

Yet, Craigslist just scratches the surface of what is possible with local classifieds because it only offers raw information. It doesn't provide research. As we wrote in our overview of the classifieds market several startups are creating vertical applications that not only show listings but augment them with research. For example, Trulia is a vertical search and research tool for real estate.

Another example of enhancing classifieds is adding salary information. The job search engine Indeed shows not only available jobs, but local and nationwide salaries for the exact job you are searching. Recognizing that classifieds are tightly coupled to research is important because most decisions, like getting a new job or buying a new house, are not made on a whim.

Hyperlocal Social Networking?

One of the big pet peeves of American culture is privacy. Arguing with your neighbors about who would pay for the fence between your houses was the thing to do in the 1950s, 60s and maybe even in the 80s. Are people ready to warm up to their neighbors? StreetAdvisor and FatDoor think that the answer is yes. Both startups offer social networking based around your local neighborhood. While FatDoor is still in the stealth mode, StreetAdvisor is up and running. The angle that StreetAdvisor uses is asking people to rate their street as a segue into local social networking.

The rating and information approach is certainly interesting, but it is still questionable whether people who can't bring themselves to say hello to each other in person, will want to first connect online. The Alexa traffic chart for StreetAdvisor hints that the answer is no, but perhaps there is a spin on this that might just work.

Aggeregating Hyperlocal Information

Even though people might not want to network with their neighbors, there is little doubt that people want to know what is going on in their neighborhood. The challenge of re-aggregating web information from the geographical perspective is big. Given that very little of the content is probably geotagged, sites like Outside.in and PlaceBlogger are re-crawling the web to make this information available.

The picture above shows a typical way in which local information is presented. In this case, it is from an Outside.in entry about the increasing fares of New Jersey transit. The information is aggregated from various sources and automatically tagged. The chart shows the amount of buzz on this topic in the past few days. The related links allow readers to explore more on the topic from local blogs and newspapers.

Monetizing Hyperlocal Information

The race to build businesses around hyperlocal information is not accidental. There is a belief that there's a big monetization play here. In our recent conversation, Outside.in CEO Steve Johnson told me that knowledge of the geography of the audience is very important to advertisers. He cited two examples. The first, was the current presidential election. Parties could send direct advertising to zip codes that are known to be tipping points in swing states. Since the stakes are huge in each election, each party would pay big premiums to reach undecided voters on web pages with hyperlocal content.

The second example he gave was local advertising. For a while, white pages made a business by selling ads. Hyperlocal web sites, however, promise to take the concept to a whole new level by enabling local businesses to deliver highly targeted and relevant advertising. But this isn't a simple problem. To make it work, one needs to build an advertising platform where local entities can bid on ads. The platform would need to make money on the volume the ads, since doing door to door sales in such a business is not scalable.

Conclusion

The global grid is becoming ubiquitous. It is growing at a mind boggling speed and absorbing enormous amounts of information. With the spread of the grid we are observing the rise of hyperlocal information. User generated media, classifieds, and other local content is increasingly more tied to specific geography.

Despite globalization, hyperlocal information is very valuable both to people and advertisers. In the coming years, we will be seeing the rise of a new way to look at information - geography. Inspired by utility and the promise of hyperlocal advertising, startups are racing to build businesses that deliver highly relevant, local information to users.

Hyperlocal information and aggregation is just in its infancy with much more to come. In the mean time, please tell what hyperlocal information you find most interesting? What information services would you like to see around hyperlocality in the future?