Geo-targeting: fact and fiction


Bob McCurdyAdvertisers are always looking for ways to enhance the impact of their commercial messaging and with smartphone penetration approaching 70%, not surprisingly, mobile geo-targeting is generating increased scrutiny and interest. Being able to reach consumers within a pre-determined vicinity with a call-to-action offer is a grail-like goal. But at this juncture, this goal is not without challenges as it’s simply not as easy as it sounds nor as accurate as often assumed.

There are several ways location based data is generated, some of which are more precise than others:

IP Address: Accounts for a considerable percentage of all location based data. Targeting in this fashion runs the gamut from accurate to inaccurate, more often the later.

Wi-Fi triangulation: –This is extremely accurate and involves sharing of Wi-Fi location data that can pinpoint location to within a 100 meter radius.

A-GPS: This data is typically represented by latitude and longitude coordinates and is extremely accurate to within several feet. Currently, having access to this data is the exception rather than the rule as only a small percentage of mobile ad impressions are delivered in this fashion.

Cell Towers: Can be accurate when enough towers are available, otherwise location projections can range over many miles.

Registration data: Vigilant steps need to be taken to maintain accuracy. Targeting via registration data is often accomplished by pinpointing the center of the geographic area or zip code that’s been supplied. The accuracy of this type of targeting doesn’t match A-GPS or Wi-Fi triangulation.

Regardless of the manner in which the location data is generated, publishers and ad networks often transform it into latitude and longitude coordinates, thus giving it the appearance of being more granular and thus more accurate than it actually is. A vendor might state that their data is latitude and longitude based but it doesn’t mean it matches A-GPS or Wi-Fi triangulation precision.

Telenav recently recently shed some light on the state of Mobile geo-targeting with the publication of their second quarter 2014, “Mobile Advertising’s Guide to Location Accuracy” where they concluded that:

34% of Mobile geo-targeted impressions are accurate within 328 feet of the consumer’s actual location.

9% of Mobile geo-targeted impressions are delivered within 329 feet- 2/3’s of a mile of the consumer’s actual location.

30% of Mobile geo-targeted impressions are delivered within 2/3’s of a mile- 6.2 miles of the consumer’s actual location.

20% of Mobile geo-targeted impressions are delivered within 6.2 miles- 62 miles of the consumer’s actual location.

7% of Mobile geo-targeted impressions are delivered 62 miles+ from the consumer’s actual location.

So per Telenav, about one third of all mobile impressions are delivered within a football field of the location with 43% delivered within a 2/3 of a mile radius. Not bad but not as accurate as often thought. Verve CEO Tom Maclsaac once wrote, “There’s one very significant problem that is holding mobile location targeting back. The market has been flooded with bad location data. And many are turning a blind eye to this fact because to acknowledge it would be contrary to their business interests.”

The Mobile industry’s geo-targeting capabilities are “not there yet”. But without question its accuracy will continue to improve, which is why it’s important for the Radio industry to continue to vet various technologies and options that will enable Radio to keep pace. A couple of geo- targeting options of which I am familiar are offered by Clip Interactive and NextRadio but I’m sure there are others. For broadcast Radio to retain its relevance and grow revenue, it’s imperative that the industry be able to go shot-for-shot with the Mobile industry when it comes to its targeting capabilities for both its digital streams and analog broadcasts.

–Bob McCurdy, [email protected]