Braze has entered into a definitive agreement to acquire OfferFit, an AI decisioning company, for US$325 million, subject to customary closing adjustments.
In September 2024, Braze shared its vision for agentic AI in customer engagement, building on its existing artificial intelligence research with the planned development of Project Catalyst.
This native AI agent is designed to help brands personalise and optimise experiences with highly relevant Journeys, Content, and Incentives.
For four years, OfferFit has been building a multi-agent solution that recommends individualised cross-channel customer journeys in partnership with customer engagement solutions like Braze.
Already brands have seen improved uplift by using the OfferFit AI technology, for example personalising over a hundred characteristics within an email to promote new customer signups, or optimizing reactivation campaigns for inactive users.
Once closed, Braze will be able to both more deeply integrate the existing sophisticated, multi-agent approach from OfferFit into its broader platform.
Braze will also be able to leverage expertise from the OfferFit team alongside current Project Catalyst development to collectively drive more sophisticated experimentation and deliver 1:1 personalisation for all aspects of a customer experience.
“Combined with Canvas as an orchestration layer and infused with BrazeAI throughout our stack, Braze is now responsible for the rapid and reliable delivery of trillions of personalised messages each year,” said Bill Magnuson, co-founder and CEO of Braze.
“And now, with Braze’s planned acquisition of OfferFit, our complementary products and teams will come together to define the next chapter of the evolution of Customer Engagement with AI,” said Magnuson.
George Khachatryan, co-founder and CEO of OfferFit, said OfferFit was built to apply advanced technology to the hardest problems that marketers face.
“Focused on the decisioning layer, OfferFit replaces the manual work of A/B testing with reinforcement learning agents that autonomously experiment and learn optimal actions,” said Khachatryan.
