Skip to content
News Company news

Essence scales use of advanced machine learning with Olive integration

Essence comms icon square


Since the beginning of the year, Essence has been pioneering the use of machine learning to power bid-optimization in Olive, our proprietary media management software. 

The integration makes it easy for Essence employees and clients to capitalize on the power of machine learning to drive campaign performance at scale by combining Olive’s easy to use interface with scoring mechanisms developed by Essence’s analytics and data strategy teams.

Integration of AI-enhanced bid optimization into Olive accelerates programmatic optimization capabilities and makes machine learning easy to deploy across campaigns. 

“Machine learning is helping us automate and accelerate tasks across almost every part of the campaign lifecycle," says Essence cofounder and Chief Product Officer Andrew Shebbeare. "What's exciting about this development is that by productizing the interface between humans and machines, we are now able to put the power of machine learning in the hands of as many people as possible - and deploy it as broadly as possible - without the need for any additional technical expertise.”

With the integration of the feature into Olive, this enhanced service can now be deployed on programmatic campaigns by any Olive user with just a few clicks, delivering continuously increasing cost efficiency and incremental brand lift improvements. Users can select from a range of proven scoring mechanisms designed to enhance a wide variety of campaign objectives and strategies. 

Essence began testing bespoke bid optimization tools with the aim of improving programmatic campaign efficiency and impact on brand awareness. In the process, the agency developed a range of impression-scoring formulae to guide automated, impression-level decision making. Essence’s bespoke formulae create automated, tailored, and self-improving bidding strategies that ensure the agency’s clients are able to access the most valuable impressions at the most efficient price point.

During testing, campaigns utilizing this new machine learning-enhanced approach outperformed all other approaches, achieving improvements in brand lift over and above well performing control initiatives. Impression share on premium inventory more than tripled while cost per viewable impression fell by 34% compared to control initiatives, without any negative impact on scale.

For more about how we're applying the power of AI to advertising campaigns, read "How WPP’s Essence Uses Machine Learning to Improve Media Buying Results: Creating an agency network fit for the contemporary marketplace" in Adweek.