American Marketer


Mobile display ads: Early data shows that double opt-in does not improve quality

April 16, 2014

John Busby is senior vice president of the Marchex Institute John Busby is senior vice president of the Marchex Institute


By John Busby

Over the last few years, mobile marketers have complained about accidental clicks from mobile display campaigns. Who else remembers experiences like being inadvertently taken to the App Store in the middle of a Words With Friends match?

Studies have confirmed this issue, suggesting in some cases that more than 95 percent of interactions with a mobile display banner are unintended.

The ad tech community has responded, introducing a number of new products designed to confirm that the mobile user would like to follow-through with the ad.

These products often include a swipe or secondary action, hence referred to as “double opt-ins” or “double opts.” They are great ideas, and were designed with both the consumer and advertiser in mind.

So, do they work?

Double up
Early data shows that double opt-in has only a modest impact on accidental clicks. It has not improved the overall quality of ad campaigns enough for us to invest considerably in marketing products that are high-consideration or require a higher degree of consumer buy-in than an app download or listicle.

Back in 2012, we found after analyzing six mobile display networks that it took, on average, 2,481 clicks to produce 29 phone calls (these were click-to-call campaigns), and that these 29 phone calls produced only one “high quality” phone call, meaning it was from someone who really wanted to talk to an advertiser about their product or service.

What were the other 28 calls? We refer to them affectionately as “baby calls,” because the data showed us many of these calls were clearly made by babies playing on their parents mobile phones. If you have a four-year old, as I do, this is probably an unsurprising finding. The rest of the calls were a mix of accidental calls and spam calls. So how about in 2014?

Millennials may be digitally-savvy, but post-millennials are digital savants. It turns out that babies know how to bypass the double opts, as well.

This year it took slightly more than 1,200 clicks to produce the same 29 phone calls, and those 29 phone calls produced three “high quality” phone calls. Better than before, but not even in the ballpark compared to other digital channels, in which we expected most clicks we paid for to turn into phone calls. And, of those, we expected up to 40 percent to be “high quality” calls.

So mobile display still has some serious catching up to do, and despite our best efforts in the ad tech industry, it may just be really difficult to expect a small banner to produce big results.

Make the call
The antidote for marketers is to focus only on campaigns where the results can be inarguably measured. For high-consideration products, a coupon, call-tracking number or a simple form are good starting points.

For brand marketers, focus on application downloads or some type of user engagement.

If you are dead set on an awareness campaign, run some tests with engagement elements to get a baseline of which networks or publishers will perform best for you on a relative basis.

And by all means, do not let your baby see your lock-screen password.

John Busby is senior vice president of the Marchex Institute, a Seattle-based research and analytics entity that publishes findings on mobile advertising and the digital call advertising industry. Reach him at