How Watson AI allows IBM's Advertising Accelerator to do three key things

At CES 2020, IBM launched its Advertising Accelerator with Watson. IBM's Dave Neway describes the benefits this solution offers to the advertising community.

How Watson AI allows IBM's Advertising Accelerator to do three key things At CES 2020, IBM launched its Advertising Accelerator with Watson. IBM's Dave Neway describes the benefits this solution offers to the advertising community.

TechRepublic's Teena Maddox talked to Dave Neway, head of marketing at IBM Watson Advertising, at CES 2020 about how Watson can take advertising to the next level with artificial intelligence (AI). The following is an edited transcript of their conversation. 

Dave Neway: We had a major product announcement this week. On Tuesday, I'm very proud to announce that we launched a new offering called Advertising Accelerator with Watson, which as you can probably tell by the name, leverages Watson AI really to do three things.

Anticipation: It helps advertisers better predict as opposed to react to the optimal combination of visual elements to drive the highest engagement for an audience. It also helps in segmentation and really sort of uncovering the sort of complexion and preferences of audiences as they respond to advertisements. And it also helps in revelation and revealing the insights that help sort of drive this higher level of engagement. So anticipation, segmentation, and revelation--that's really what this new product brings to the advertising community.

SEE: IBM points Watson at the marketing world with Advertising Accelerator (TechRepublic)

Teena Maddox: That's really fascinating. Tell me a little bit about the tech behind that and how this is being measured and how you develop that.

Dave Neway: When we started developing this solution, we really tried to take a very customer-centric approach and sort of understand what the pain points were amongst our core buyers, which is kind of fourfold.

One, I think converting audiences is a very tall order. I think there's a lot of banner blindness in our industry, and our buyers are looking for solutions that really help audiences take that action. Also, in this world of personalization, sort of driving conversion, there are tactics that are available like SEM [Search Engine Marketing] and retargeting and DCO [Dynamic Creative Optimization], but that market's pretty mature. It's been sort of 10 years since those solutions have been in market. I think it's kind of time for that sort of bar to be accelerated, if you will, and sort of be enhanced.

And I think also, our buyers are looking to get their toe sort of dipped into AI, but they want to do that with AI they can trust. And that's really where Watson comes in to sort of provide that credibility and authenticity in terms of the coding that we're using.

Teena Maddox: And with this particular use of data analytics, I know that a lot of companies have been collecting data on their customers, but they don't really know how to use it. Is this a good way to find an application?

Dave Neway: That's a great question. One of the great things about this new product, about Advertising Accelerator, is that, whereas other solutions like DCO are sort of wholly reliant upon context and behavioral information and cookies and the psychographics of audiences, and we know that cookies are becoming more elusive in this privacy-safe world, this solution actually doesn't rely on context at all. It just really relies on hypothesis testing and serving ads and sort of understanding what is resonating from a visual standpoint with various audiences, and then predicting what permutation of all the various elements you can see in an ad. Is it the foreground, the background? Is it the actor that's featured? Is it the call to action that's going to make you take an action versus me, versus anyone else? It really sort of personalizes, in an anonymous way, this sort of ad experience for audiences at scale.

SEE: CES 2020: The big trends for business (ZDNet/TechRepublic special feature)

Teena Maddox: And then how does it also deal with customer privacy? Because I know that's huge.

Dave Neway: The good thing about this is that it doesn't require any sort of PII [personally identifiable information] in order to sort of serve an ad to a consumer. It really sort of anonymizes the process and serves ads just based on whether the user is clicking on the ad or not. And then it also takes into account what were the visual elements that maybe made that user click. Was it the orange background? Was it the purple call to action? Was it the use of a certain sort of weather situation that was featured? So it really kind of, sort of democratizes, how performance creative works in this new world.

Teena Maddox: I love that. So they can figure out what color is working, what headlines, even what font.

Dave Neway: And I think the benefit of that is that it kind of flips the script a little bit on sort of strategy informing media. I think with a solution like this, the media actually informs the strategy. The insights that we derive from running a campaign through Accelerator would reveal that, "Okay, it looks like there's an audience here that is really responding well in the Miami demographic to orange backgrounds with this type of CTA." Whereas, maybe in Milwaukee they're responding to trees in the background and an orange or a purple background with a completely different call to action. It really kind of helps advertisers better understand the complexion of their audience and shape their future media strategies accordingly.

Teena Maddox: Fascinating. And when will this be available?

Dave Neway: It was made live, actually on Tuesday of this week. It'll be announced with two beta partners as well: LendingTree and the Potential Energy Coalition. And CES has been wonderful in terms of generating more interest. [There's been] a lot of hand-raisers, and we're very excited about the road ahead.

Also see

20200110-ibm-dave-teena.jpg

IBM's Dave Neway at CES 2020 talking about Advertising Accelerator with Watson.

Image: TechRepublic