There is a lot of talk about machine learning in marketing these days, and a lot of vendors beginning to offer the capability.

But how does ML work in advertising, and what can be gained?

In this video interview with Beet.TV, Jim Lawson, CEO of one vendor offering the technology, explains.

“Whatever your business goal is, if you have enough data and there is a ton of data out there, the question is there’s too much data, what do you do with it?,” says Lawson of AdTheorent.

“We take the data, we integrate them into our machine learning models and algorithms and then we predictably score every impression that comes into our system.

“We serve ads to the high predictive scores, and we ignore the low predictive scores all the while optimising the best price.”

Eight-year-old AdTheorent offers predictive targeting, geo-targeting, audience relationship graphing and creative production.

Lawson says “science” can solve advertising’s persistent waste problem.

There’s a lot of waste in digital advertising, so we use science … it is consistently and daily executing on campaign objectives for your clients as measured by their goals, not by clicks.

This video is part of a series of interviews conducted during Advertising Week New York, 2019.  This series is co-production of Beet.TV and Advertising Week.   The series is sponsored by Roundel, a Target company.  Please see more videos from Advertising Week right here