Is Retargeting the New Pop-up?

This stems from an uncomfortable truth. For all the billions poured into advertising technology over the last decade, the one sure-fire hit has been retargeting, a practice that’s been around since the 1990s when DoubleClick introduced “boomerang.” The difference now: exchanges offer deep pools of cheap ad inventory to match with cookies of store abandoners. Since top retargeting providers operated on a pay-per-click model, the method is right in the wheelhouse of direct-response budgets.

The problem is too much of a good thing is a bad thing. Sure, retargeting works. People often leave shopping carts and then return. It makes intuitive sense to remind them, maybe ply them with a discount offer. But there’s that undefinable creepy factor that most ad targeting skirts right up against.

“I’ve heard about a game people play where they get their friends to click a URL for a specific ecommerce item simply so it’ll follow them around the Internet,” said James Green, CEO of retargeting firm Magnetic. “Clearly that’s an issue.”

The problem is not so much around the concept of retargeting as it is the execution. Many advertisers making use of the practice currently do so using multiple providers on a single plan. That means frequency caps go out the window completely. One firm might limit the number of impressions a user is exposed to to five, for example, but there could be five or six out there doing the same thing. The result is often that users are overloaded with ads for a brand who’s site they visited briefly, often with no intention of actually making a purchase.

“It’s a matter of sophistication, and that will evolve drastically,” Green said. “There’s still a bunch of people out there serving ads to anyone that hits their site. Often people are retargeted for weeks after they’ve already bought products. There’s clearly a lot of wastage out there.”

But despite that waste advertisers still see upside, so continue to dial up their investment in it. Retargeting has seen a huge uptick in adoption over the past few years, thanks to dedicated ad networks and, of course, the rise of exchanges. Exchange operators and DSPs estimate as many as 50 percent of all ad impressions purchased through exchanges are informed by first-party targeting data, read: retargeting. Even on Facebook a large portion of ads are now retargeted. In fact, Facebook’s ad exchange took off on the back of retargeting. Don’t be surprised if Twitter is next. Retargeting is big business. Top provider Criteo has over 800 people working for it. The company is widely seen as a top candidate for an initial public offering.

That begs the question of whether or not there’s a bubble forming around the practice. It’s easy for retargeting vendors to tell a compelling story, since they’re often the ones responsible for the last-click or last ad view before a user returns to a marketer’s site.

Just because a user returns to an ecommerce site after seeing the shoes they’ve already decided they’re going to purchase in a retargeted ad, however, that doesn’t mean top of-of-the funnel activity had no influence on their behavior. Retargeters often starting on third base. As marketers figure out better attribution methods retargeting won’t get the same credit it does now, and that could see some dialing down their spend on it.

“There’s no doubt that as marketers get more sophisticated about upper-funnel attribution they’ll cut back on some of their basic retargeting spend,” Green said.

In the interim marketers will just keep doing what they see working, regardless of murky attribution models. Those seeing their increase in retargeting budgets directly correlate to increased sales would arguably be stupid not to, despite the fact they’re buying impressions they don’t need to. If chasing users round the Web for weeks shifts product, they won’t stop any time soon, unless the numbers shake out and they can prove to themselves that retargeting is actually standing on the back of other media and calling itself tall.

The industry just needs to hope that consumers don’t become so frustrated with seeing the same pair of shoes for two weeks that they decide to do something about it. Otherwise it’ll only have itself to blame.