Creative bid optimization requires data and the ability to predict not just the intent of the consumer to make a purchase, but when that purchase will occur. Serve the ad too soon and the brand can miss the opportunity. Serve the ad after the purchase and the brand risks losing the customer for life.
On Tuesday, IgnitionOne announced full support for Google Shopping campaigns through its Digital Marketing Suite (DMS). Rather than the features touting campaign management, feed integration, analytics and support, three little words --predictive bid optimization -- caught my attention.
So I asked IgnitionOne President Roger Barnette his thoughts on whether Apple, Google and Microsoft would use the artificial intelligence in Siri, Now and Cortana to build out predictive ad targeting based on their respective technology's AI.
Barnette said Siri and Cortana have specifically created AI for the purpose of emulating an executive assistant on your smartphone. IgnitionOne's AI is made to optimize the media-buying process. The use of AI will continue to grow -- but for specific reasons, Siri will not ever manage bids for you in Google, he said. But I think a combination of Siri's technology, business intelligence and analytics will support this theory to improve ad targeting on devices running Apple's operating system (OS).
Microsoft Cortana and Google Now, respectively, will work similarly on mobile devices. The problem of inaccurate ad targeting would become a thing of the past if a version of these AI systems tied to personal app features and business enterprise systems is developed. This would be the result of an integration of data from services like email, search, social networks, and calendar entries, along with cross-device targeting on desktop, television, set-top boxes, tablets and smartphones -- and CRM platforms, inventory control, and materials replenishment. Location-based services and attribution will also play an important role. These developments will send brands to nirvana by bringing down ad targeting costs while improving accuracy. (I'm sure most of you think I'm living in a fantasy world, but we're still years away from making this a reality.) The technology exists -- but someone needs to figure out how to put it together.
Barnette takes a different point of view. "Rather than integrate with those types of artificial intelligence, I think we will see predictive analytics becoming more powerful over time and will instead mirror what we think of as AI," he said. "I also think it's important to note that no matter how intelligent systems become, there will always be a human element. Marketing is art as well as science -- and it is our job to make the science side as simple as possible to let humans create their art."
Leadspace CEO Doug Bewsher, former CMO of Salesforce, said predictive analytics also means understanding people rather than behavior. "We will see a transformation on how it all connects, but people have not quite figured out how to use the amount of data that exists today," he said. "It's very early days when it comes to figuring out how marketers can give consumers just what they need."