The Trickster: Data's Sleight-of-Hand Danger

by Chuck Moxley, CMO "audience targeting", data, location, "store visits"
The Trickster: Data's Sleight-of-Hand Danger

Data, like a magician, can distract you from looking where you should be looking. Problems with bad data are common and can mislead your mobile advertising and campaign measurement efforts.  

Data Distraction – one of five big Location fails made by mobile advertisers – happens when you make the mistake of assuming that the location information you’re getting from data is accurate. It’s a well-publicized fact that a lot of the location data from ad requests isn’t accurate. Recent research showed that nearly two thirds of ad requests do not contain sufficiently accurate location data to target a user.

If you are not validating your mobile campaign data’s location accuracy, or at least working with a platform that does, you are making assumptions that will result in problems such as bad targeting and inaccurate contextual messaging.

A sampling of reasons why the location data in ad requests is often wrong:

  • If the user doesn’t have GPS turned on, the only source the app can use to determine location is IP Address – a notoriously inaccurate way of tying to a physical address, with a fail rate greater than 25 percent.
  • Some apps don’t obtain device location, instead deriving it from data supplied by the user when they registered the app.
  • To mitigate battery use, some apps only report the last location ping the phone generated, caching the result. As soon as the device moves, the actual location data becomes false.
  • Some apps intentionally obscure location accuracy for privacy reasons.
  • Some app makers obscure lat/long data to sell it through their own sales force.
  • The phone itself may be inaccurate, such as when the device user is indoors and the roof obscures the GPS signal.
So, why does location accuracy really matter to you as a marketer? Three reasons:
  • If you are using store visits to measure campaign effectiveness, you may be assigning store visits when the visit didn’t actually happen, or you may be missing store visits that actually occurred.
  • Location reliance could result in false assumptions if you’re developing location-based audiences. If the location is wrong, you are assigning false information to the device history algorithms that could change the interpretation of that individual. Are they at a coffee shop or gym or donut shop? The difference radically changes which one you want to target.
  • If you are using geofencing, you could be geofencing the wrong person and missing the right one, such as when someone is in or near a store. If the device location data is off, you’re wasting costly impressions by targeting customers that are nowhere near you’re store.
Bottom line: if you’re going to rely on location in mobile advertising, then you’d better be certain you’re using accurate location data.

To learn more about the risks of becoming a victim of data’s sleight of hand, and to see three best practices to reduce your data reliance risks, download a free copy of the whitepaper Location Fails: Five Major Mistakes Made by Mobile Advertisers. In it, Jeff Xouris from WPP agency VML and I reveal the five most common mistakes marketers make when leveraging location in mobile advertising, and how to fix them.