• by Erik Sass  @eriksass1, May 19, 2014

Location data is the “killer app” of mobile advertising and marketing, but a surprisingly large amount of the information used to target mobile users based on their whereabouts is imprecise or just plain wrong. However, mobile ad platform Thinknear is hoping to raise awareness of this issue with Location Score, a new index that measures the overall accuracy of location data used in mobile advertising.
 
Thinknear founder and general manager Eli Portnoy explained that an accuracy index of this kind for location data would have been impossible even a few years ago, when only about 10% of the ad inventory had latitude and longitude information attached.

Today, that proportion has risen to 67%. But Portnoy added: “The bad news is that the quality of that location data has degraded considerably over time. We’ve been working over the last two years to accurately score location and only buy location data that’s accurate.”
 
To create the Location Score index, Thinknear “looked at 3.5 billion ad auctions on exchanges and bought 53 million ad impressions. Then we asked the users if we can pull their location using GPS, and compared it to where the ad network said they were.”

According to Portnoy, this should allow mobile marketers to compare location-based ad impressions to actual location data to find out just how accurately their ads are being delivered.
 
The current answer is “not particularly accurately.” Portnoy said the Thinknear Location Score index for mobile marketing overall, based on the first round of measurement, is just 49 out of 100. Taking a closer look at the numbers that go into that score, “35% of the impressions are accurate within 100 meters, 9% were between 100 meters and 1000 meters, and 30% were between 1,000 meters and 10,000 meters,” which Thinknear terms “regional” inventory.
 
Finally, 19% -- nearly one-fifth of mobile location-based ad inventory -- was outside 10,000 meters, meaning more than six miles off target. In everyday geographic terms, that means an advertiser might target your mobile device with ads for businesses in Times Square, when you’re actually visiting the Statue of Liberty.
 
On the positive side, the promise of contextual location-based targeting is real: Portnoy said Thinknear is “finding that better location data quality correlates very directly and linearly with conversion” as tracked by various metrics, including time spent on site, traffic to an advertiser destination online and visits to brick-and-mortar retail establishments.  

advertisement

advertisement