• by Steve Smith , Staff Writer @popeyesm, May 7, 2015

The purported ineffectiveness of mobile banner ads has become such an article of faith among many marketers and agency execs, its offhanded mention barely gets contested anymore. But whether we like them or not (and “we” don’t), these diminutive eyesores of mobile can have a cumulative impact on viewers -- an impact that's measurable. These ads even seem able to influence consumers' retail decisions. Go figure.

In a new series of benchmarks derived from over 83 mobile campaigns run on Catalina’s ad platform powered by 4Info, the companies claim their advertisers saw a $2.57 return in retails sales on every dollar spent on mobile ads. The attribution was done by associating user phones with specific households and then using Nielsen Catalina Solutions data to determine sales lift for users exposed to the mobile ads. These figures represent a 15% better impact on sales than desktop Web ad campaigns. 

And according to 4Info CEO Tim Jenkins, “87% of the campaigns were standard banner ads.” With the right mixture of targeting, creative and frequency, even the lowly mobile banner can move the needle on offline sales, if these numbers are accurate.

Given the ties to Nielsen’s Homescan and the Catalina platform, most of the campaigns involved CPG categories. According to these results, the campaigns resulted in an average sales lift of $29.90 per 1,000 impressions and a half million in sales per campaign. Of course, 4Info argues that they aren't serving just any old mobile banner, but one that is targeted down to the household level and against past shopping behaviors. The full study is available here. 

Promotional and coupon campaigns far and away drove the most sales lift. But creative and frequency also proved important. Jenkins says that for a family familiar with a brand or into the category, 5.7 banner exposures optimized retail spend. “But when you start to deliver 10 or 11 impressions, you are now generating less revenue than if they saw only one impression,” he says.

With video, you can cut those frequency numbers in half. Streaming mobile ads generate 4x to 5x in store sales as banner ads. Fewer impressions will generate the same level of sales lift. At the same time, video can easily alienate viewers if seen too frequently.

The attribution system 4Info uses, which relies on actual store sales, poses an interesting challenge to campaign optimization. There is a necessary lag between a campaign launch and the reporting of results, says 4Info CMO Chuck Moxley. “If you have an 11-week campaign, typically there will be a period of four weeks after to count sales and allow people to see the ad.” Add to that the three weeks it takes to complete the study and “it is a challenge for in-flight optimization,” he admits.

4Info is trying out an approach that allows weekly reads on the more basic in-store sales statistics so that advertisers might optimize against factors like time of day, creative, and whether the user was seeing the ad at home or away. They are also adding store-visit metrics, which are a very rough but more real-time indicator of activity that might have been activated by a mobile campaign.

And so, if 4Info’s formulas are right, we may all owe the mobile banner an apology for all of the contempt we have been heaping upon it.

Don’t all rush in, now.