By Julie Polmonari, Waze Northeast Sales Lead
We marketers know that location-based marketing is big—so big that, as GeoMarketing notes, industry experts suggest the space will grow to $18.2 billion in advertising revenue by 2019. The very name “location-based marketing” would lead anyone to conclude that companies offering location-based services are doing so with a high degree of location accuracy. If a key value is the technology behind understanding a person's location, couldn’t one assume that technology is consistent and precise? Unfortunately, that's not always true.
It may seem ironic that many location-based media services are inaccurate. In fact, a whopping 54 percent of “location-targeted” mobile ads are off by more than a half a mile, according to a Thinknear study. In practice, this means that a coffee chain can think they are targeting a mobile user within a few blocks of a store, making it easy for that person to stop in on their way to work—but in reality, that person is on the other side of town. How can these inaccuracies persist? Let's look at the targeting methods used by publishers, as defined by the Mobile Marketing Association:
Cell Tower Data
Cell tower triangulation determines a user's location based on the strength of the signal between the device and nearby cell towers. The stronger the signal, the closer the device, so using three cell towers (triangulation) can determine the approximate location of the device. Utility depends on the density of cell towers in a given area, making the location accuracy of this method less reliable in suburban and rural areas.
There are a few ways to use Wi-Fi to determine a user's location. Wi-Fi location is derived in the same fundamental way that location is derived with cell towers, they’re simply smaller, therefore more precise. Triangulation between multiple Wi-Fi signals is again possible, but latency can be an issue as the processing time for determining a location is more complex. This targeting method is popular since it seemingly requires nothing but a phone with Wi-Fi turned on—however, you also need a user to opt in to an app to share their Wi-Fi signal externally. There have been recent concerns related to privacy and tracking without permission; mobile advertising company inMobi was recently fined $4 million dollars by the FTC for tracking users' Wi-Fi signals without their permission.
GPS is the most accurate of all geo-location services and is viewed as the "gold standard" in today's marketplace. GPS isn't just used for ad targeting, it's also used in other capacities that require pinpoint location accuracy. For example, it's used in navigation systems (sea, land and air) to provide accurate directions. A GPS signal is actively collected when a user is in-app, or less accurately when an app is turned on in the background (after accepting permissions). GPS is incredibly accurate, with the ability to determine a location based on latitude and longitude coordinates. However, it depends on a user opting in to allow GPS targeting—making scale a concern.
Why is accuracy important? By missing a user's location, marketers unknowingly miss their audience. As an example, one of my San Francisco-based colleagues saw a mobile banner ad for a fast-food chain while browsing in-app. Out of curiosity, she mapped the location showcased in the banner—only to find it was a restaurant in Washington state. While she had been in Washington state weeks before, there's little any restaurant could offer to encourage her (or any user) to drive more than 13 hours to the promoted location.
So, what's a marketer to do? Understand the way your media partners arrive at location. Know which methodologies they employ, understand the benefits and drawbacks of each, and ensure they have the scale to execute your program. Customers don't want to drive 13 hours out of their way to act on an ad, and marketers deserve assurance that precision is part of the deal.
This article was originally published in The Compass- an industry resource for mobile, native, and location-based marketing.