Today, I was looking through my server statistics for an account at one of the hosting providers I maintain, and I noticed something pretty funny and incredibly timely, considering everything that has been going on in the world as of late.  It seems that I’ve been getting a lot of traffic from Yahoo Local from searches for “gas prices in Chicago”.  This brings back some good memories:

It was about three years ago that someone forwarded me a chain email that was a plan to fight high gas prices.  The idea was to boycott Exxon Mobil.  The email claimed that by doing so, it would force the world’s largest oil company to react by lowering their prices, in order to “drive more traffic” to their stations.  The smaller competitors, faced with increased price competition from the giant corporation, would have no choice other than to respond by lowering their own prices.  Although completely unrealistic, the theory actually made some sense.

At the time I read this email, I was part of a small internet startup called LocalLaunch.  We focused almost exclusively on local internet marketing.  At that time, this was a specialization which very few companies had.  So, I spent a lot of time thinking about how to generate internet traffic from geographically qualified searches.

One of the early search properties to launch specifically serving the local small business market was Yahoo Local.  Yahoo Local integrated a directory taxonomy with very basic social networking tools (in the form of user reviews).  The earliest version ranked its results based on a fairly crude algorithm which calculated relevancy, ratings, and distance. 

Herein lies one of the biggest challenges with truly local targeting on the internet.  Every connection to the internet is provided with a unique IP address that can be used to indicate the location of the user, so it very often gets used in the targeting of search advertising results.  When a user searches for “gas stations” with no further geo-qualification, it makes sense that they intended to find a business close to their location.  IP targeting works fairly well when covering a large area, but it becomes significantly less accurate when the target is localized to a specific part of a city, neighborhood, or a radius.  This is because IP addresses are provided by internet service providers (ISPs).  Depending on where you live and what web service you use, your provider might not be hosted locally.  The result is that your IP address may indicate that you are sitting far from where you are actually conducting your searches.

So, IP targeting does not work that well for true local plays, such as Yahoo Local.

Unlike the PPC engines (Google Adwords, MSN Adcenter, and Yahoo Panama); Yahoo Local does not use IP targeting to calculate the distance portion of it’s ranking algorithm.  Instead, it measures the distance between a business address and the “center of a city”.  The city center is calculated by doing a scattergram of all business addresses within a given city, and measuring where the business density is highest.  In the case of Chicago, that location is slightly West of downtown.  This is also done for zip codes, neighborhoods, etc.

This method of determining distance works pretty well, really.  A user searching Yahoo Local for “gas prices in Chicago” will be given results that are downtown, and a user searching for “gas prices in 60601″ will be given results that are in 60601.  Most of the time.

Yahoo Local has never been as hard to get into as Google Maps.  It is, in fact, quite easy.  Additionally, also unlike Google Maps, address verification has never been much of a priority.  So, when I received the chain email with the silly plan to fight high gas prices, it gave me the idea for a test.

I built www.neoactivism.com to test whether a site with no affiliation to an address could rank in Yahoo Local. 

I used the content of the forwarded email as my content.  I conducted my own estimation of the city center in order to pick an address to submit for my listing.  Guess what?  It worked.  I started ranking #1 in Yahoo Local for “gas prices” in “chicago”.  The experiment was a success!

That was before gas prices topped $4 a gallon, and they became one of the hottest topics in America.  Now, I am getting tons of traffic from my listing.  And, aside from some poorly constructed Adsense and Amazon affiliate links, the site has no revenue generation stream whatsoever. 

So, let’s hope that whoever originally wrote that chain email knew what they were talking about.  If so, perhaps some good could come of this, afterall.