Expert: What Is Working, What Isn't In Local Marketing
October 14, 2015 – BizReport When geo-targeting first came on the scene it was based on a cyber ‘fence’ around a business. That fence seemed like a good idea, but just having one of these fences around a business didn’t identify audience members. One expert tells us how to better use geo-targeting.
Kristina: Geo-targeting/location targeting has been around for a few years now. What is working and what isn’t with the current crop of geo-targeting?
Michael Hayes, Chief Revenue & Marketing Officer, UberMedia: [One issue is] that the boundary geo-fence misidentifies audiences. Meaning, if you wanted to target customers at a Kia car dealership, you likely also served ads to people that were at the BMW dealership next door, those at Target down the street and anyone driving by on the highway. This resulted in a lot of wasted advertising because none of those other misidentified audiences were likely in the market to purchase a Kia.
The next evolution was to use 100 meter by 100-meter tiles and target users within a tile or a few tiles. This wasn’t much better as retailers often crossed over in multiple tiles. The tile method also misidentified audiences with other irrelevant locations, residential areas, highways, etc. After all, 100 meters is the size of a football field and not all businesses are that large.
Kristina: Are marketers still using this kind of targeting?
Michael: There are still mobile publishers/vendors that target with these sloppy systems, and the feedback we hear is that they get below average results. What works best is to identify audiences precisely based on their location context. We accomplish this with our Polygons, which are drawn precisely around a specific location, that are accurate within three feet. This solves the issue of misidentifying audiences and works incredibly well.
Kristina: Some have said geo-targeting in retailers is the best way to engage customers, but you disagree. Why is that?
Michael: If your strategy is to serve ads only to people inside a retailer, for the foreseeable future, this is a poor tactic. The reason this tactic tends to lead to underwhelming results is because some locations don’t get very much foot traffic in any given time period. For example, how many people are at a pizza restaurant during dinner? A few hundred? That’s not many ad impressions. Further, most location ad solutions can’t identify enough consumers inside a location. In short, scale is usually an issue.
What’s interesting is to identify the type of people inside a location or a competitor location, to expand that audience through advanced modeling and then re-target those users with your advertising message. We call this “real-world location behavioral targeting.”
More from Michael and UberMedia later this week, including his top 3 tips for better geo-targeting.
This article was originally published by Kristina Knight for BizReport.