Adobe updates Target with Sensei-powered enhancements

From Adobe’s blog post on the new Target enhancements

Adobe is out today with some enhancements to its Target A/B testing tool, all of which it says are made possible by its Sensei cloud-wide layer of artificial intelligence.

The tech giant released its One-Click Personalization last November in beta, which enables algorithmic page-level testing, and now it’s out in general release with some new features. These include something called Backup Policy, where no variation will perform worse than the control version you set up.

A beta update of recommendations in Target now utilizes an algorithm that Adobe said “was inspired by” natural language processing, although it doesn’t directly use NLP.

The new algo looks at all the interactions a user has had with the brand as intent signals, including choices, clicks, and views on all platforms, in order to make a recommendation.

Adobe suggests an example use case of a visitor who has watched a video on eco-friendly laundry techniques and purchased compostable dryer sheets, and its model then recommends eco-friendly detergent. Before, the recommendation might have suggested detergents that people with similar profiles also used.

There is also now automatic delivery of one-to-one offers on mobile and Internet of Things platforms, which was not previously available for those devices.

And there’s a tighter integration between Target and the data management platform capabilities within Adobe Audience Manager, enabling what it is calling Experience Versions:

Previously, Target head of product marketing Kevin Lindsay told me, there was an integration of core services between the two, but now there’s a direct integration intended to assist segmenting.

This new integration lets Target test multiple variations of a specific section of a web site or other experience against specific audience segments, in a more flexible way than before. In a post on the new enhancements, Lindsay gave an example:

… a global cookware company can develop a targeted offer on its website based on segments of people who’ve purchased a cast iron skillet in the last five months. That offer then automatically updates with the correct language or currency depending on the location of the customer, be it in the United States, France or Germany.

Finally, Adobe has told me that it is working on an industry-first “open technology that brands can use to insert their own algorithms into Adobe Target to enable more sophisticated personalization.” No word yet on when it will be available.

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.