IgnitionOne now uses AI to predict a display ad’s performance — before it runs

Marketers test a variety of factors for their ad campaigns, but one of the most difficult to test is the creative.

It’s challenging to determine if this photo, or this design, or this script is more effective than another, since individuals have as many different reactions as they have to, say, one movie compared to another.

Customer intelligence platform IgnitionOne is trying to remove some of that guesswork, releasing Wednesday a new capability that is designed to predict the performance of creative in a desktop or mobile ad, before the ad campaign is deployed.

Static or animated. “With this feature,” said COO Christopher Hansen in a statement, “you can evaluate performance before an ad even runs.”

An ad creative is uploaded to IgnitionOne’s Customer Intelligence Platform, which employs visual recognition algorithms to assess the creative’s characteristics. The creative can be a static display ad, or an animated gif.

Hansen said in an interview that the visual recognition algorithm looks for such qualities as color, image placement and kind of image (e.g., a dog or a beach scene). It then treats the specific creative combination as a single type, and deep learning AI makes a prediction of clickthrough rate for that combo for specific audiences, as defined by weather, audience segment, device, location and so on.

In the alpha test before today’s launch, Hansen claimed, the prediction accuracy as to how many users in a given audience would click on that particular creative implementation was 95 percent. No client companies are yet available to verify that figure, he said, as the alpha period used ongoing data in existing campaigns and is just now specifically launching this feature for brands.

Why you should care. Various kinds of ad platforms offer varying degrees of performance prediction, including such creative predictions as those available from dynamic creative optimization (DCO) tools. Generally, though, DCO predictions are based on the varying creatives assembled on the fly, instead of utilizing any kind of submitted creative.

But the increasing sophistication of visual recognition algorithms and pattern-detecting intelligence means that ad platforms will grow in their ability to know what approach works best for what kinds of users, such as IgnitionOne’s latest feature. The subjective factor of creative appeal becomes less subjective as the data becomes greater and the intelligence smarter.

About The Author

Barry Levine

Barry Levine covers marketing technology for Third Door Media. Previously, he covered this space as a Senior Writer for VentureBeat, and he has written about these and other tech subjects for such publications as CMSWire and NewsFactor. He founded and led the web site/unit at PBS station Thirteen/WNET; worked as an online Senior Producer/writer for Viacom; created a successful interactive game, PLAY IT BY EAR: The First CD Game; founded and led an independent film showcase, CENTER SCREEN, based at Harvard and M.I.T.; and served over five years as a consultant to the M.I.T. Media Lab. You can find him at LinkedIn, and on Twitter at xBarryLevine.