On Mobile, Google Demotes The Click

Advertisers aren't ponying up for mobile clicks. But might they pay more for a phone call or store traffic? As more of Google's users migrate to mobile devices, the company is working on new ways to measure mobile marketing success.

You click. You buy. An advertiser pays. In an over-simplified sense, that’s how desktop digital advertising works.

That system doesn’t work as well on mobile, however, where an estimated 40% of clicks are accidents (or fraudulent) and advertisers are still wary of their value. Research firm eMarketer projects that advertisers will dedicate just 2% of their budgets to mobile advertising this year—even though customers are increasingly logging in through their mobile devices.

On mobile, the simple click is failing. And it’s about to get a huge demotion.

At Google and other companies that sell advertising, the golden question has become not how to get consumers to simply click more mobile ads, but how to measure effectiveness beyond the click—even if that means tracking offline actions or purchases made on another screen.

“There’s this incredibly new, incremental engagement point called ‘out and about’ or called ‘sitting on public transportation’ or called ‘at home on the couch in front of the TV' and these are places where we didn’t used to be connected,” Jason Spero, Google's head of mobile ads for the Americas, tells Fast Company.

In these new mobile settings, maybe success for an ad doesn’t mean lots of clicks or even lots of online purchases. Maybe it means phone calls, or foot traffic to stores. Maybe it means someone searches for something now and later follows up on a desktop computer. Google has been exploring ways to measure all of these possibilities.

Take foot traffic to physical stores. With this metric in mind, a Google ad feature introduced in 2010 shows mobile users how close they are to a store location when they search for certain terms (there's also a "get directions" option). They might, for instance, search for "smartphone" and see there's a T-Mobile on the next block. There's nothing new about measuring how much these types of ads get clicked. But proving those clicks lead to in-store traffic, their desired outcome, is more difficult than simply tracking whether a click led to an online purchase.

Google is currently testing a way to measure foot traffic driven by this type of ad with five to 10 different brands. It involves comparing foot traffic in two stores located in similar cities—one that has been targeted with map data and one that serves as a control.

It aims to turn foot traffic into a measurable outcome of mobile ads, something that it has already done with phone calls. With a click-to-call ad offering, users can click a phone number within their search results to call an advertiser who has sponsored the term. Again, these clicks are easy to measure. A Google spokesperson says that on average, campaigns see on average a 6% to 8% increase in average click-through rate when brands include a click-to-call phone number in an ad. But the payoff of those clicks is less so. Google has created a product called Call Metrics to help businesses narrow in on which clicks on their ads were actually calls, call durations, and callers' area codes. Advertisers can use this information to assess how valuable calls are, and they can bid on calls the same way they bid on clicks.

Another of Google's strategies for demonstrating the value of mobile advertising beyond simple clicks has been to show, through research, that mobile searches are part of a purchase process that may end on another screen. Google released a 46-page presentation arguing this point in August. And Spero says he encourages advertising customers to build the ability to continue a mobile experience elsewhere (and the ability to track it) into their products.

In some of Google's search categories, mobile devices account for a significant amount of traffic. About 30% of restaurant searches and 25% of movie searches take place on mobile devices. About 25% of YouTube traffic is mobile. But according to earnings reports the company filed with the SEC, its cost-per-click fees and profit margins are smaller for mobile advertising products than for similar advertising on its websites. Google has an interest in evening that score.

It’s not alone. Most companies that sell digital advertising say their users migrate to mobile faster than ad dollars do. And they also have an interest in reinventing digital advertising in a way that works better within mobile context and capabilities.

Jim Payne, the CEO of mobile ad server MoPub, compares the advertising experience on mobile to that on TV. “Mobile is more of an entertainment experience [than desktop],” he says. “It’s more immersive, it’s single process, so you do one thing at a time."

He argues that it makes more sense to measure effectiveness of mobile advertising by metrics such as reach, frequency, and recall—like TV—than by the same click-through metric on which desktop digital advertising relies.

Facebook's Head of Measurement and Insights, Brad Smallwood, recently made a similar argument for all digital advertising, desktop included. He wrote in a blog post that when brands focus on reach rather than clicks on Facebook, they have 70% higher return on investment from their campaigns. Through a partnership with audience-targeting firm Datalogix, Facebook plans to provide brand marketers details on in-store sales that their Facebook campaigns ultimately generate, whether or not they were clicked on. Privacy advocates are not fans.

“The idea is to just build experiences that the mobile consumer wants, and as an industry we’re learning how to track that stuff,” Google’s Spero says.

So in other words, I ask, you can’t sell ads that you can’t prove are effective?

“We can sell them, but it works better for everyone involved when we can track and measure it.”

[Image: Flickr user Sean Molin Photography]

Add New Comment

1 Comments

  • Brad Zimmer

    How, oh how, did a marketer ever determine the worth of an advertisement before analytic data could be harvested as to it's effectiveness?  And does this analytic data give us concrete insight into effectiveness, or hints that must be reconciled in the same way an ad exec would have in the 1950s?  Maybe there's a little bit of hokum in these numbers, a way to inflate the cost of an uncertain commodity, like a futures market for screen time.