Blog

FreeWheel’s AI Approach: Built to Build On

By: David Dworin, Chief Product Officer

AI is like ice cream – it’s all about the mix-ins and toppings.

AI is everywhere—but at FreeWheel, our goal is to use it in a way that helps our clients succeed.

AI has long been built into our platform (even before it was called “AI”), a foundation built on patented technology with proven solutions. This intelligence helps us to power streaming campaigns effectively and efficiently, so it’s no surprise that we’re continuing to build new AI capabilities into the platform, too. But we’re also taking it a step further: we’re letting clients build custom AI solutions on top of our platform to deliver results that work for their business.

Think of it like ice cream: On one hand, you can go to the store and get a great flavor, but it’s pre-made, with no room for your own twist or customizations; alternatively, you can start with simple, premium ingredients and add your own toppings to create something entirely yours – endless possibilities, tailored to your taste.

In a way, that’s how we see AI. It’s not just algorithms and automation; it’s about giving advertisers and publishers the tools to innovate themselves.

Here’s a look at how we’re approaching AI at FreeWheel:

How AI is built into our platform

When it comes down to it, AI is a range of technologies that help machines replicate human intelligence – improving the way that people interact with technology and the way that different systems work with one another. In ad tech, it helps make programmatic advertising even faster and smarter—analyzing data more quickly, enabling real-time optimizations, and increasingly automating decision making. And while we call these models AI today, at the time we built them, we may have used other words like machine learning or predictive analytics.

Today, the FreeWheel platform contains dozens of AI optimizations that improve yield for publishers and performance for advertisers, including advanced algorithms to detect invalid traffic, manage pacing for live events, and optimizing pricing. We’re also regularly mixing in new models into the system that help our customers, including cost-per-acquisition based bid models, dynamic price floor optimizations, and an advanced AI-based forecasting algorithm that uses regression models on historical transactions to predict supply and demand.

In addition to optimization models, we’re also excited about the potential for agents to change the way we interact with technology, allowing us to use natural language to engage and automate parts of our workflow.

Where we’re integrating new agentic AI into our platform

Agentic AI is changing the game, and that’s why it’s a major focus in the coming months. That’s because agents help to more efficiently match buyers with sellers in a number of ways to drive better results—from helping maximize booked revenue delivery, predicting viewership, and conversion rate to enabling planning against precise audiences.

At FreeWheel, we’re developing more intelligent agentic solutions directly into our platform to help unlock a new AI era to power better connections between buyers and sellers while streamlining operations and further automating workflows.

While OpenRTB provides the rails for real-time ad transactions, agentic AI brings the intelligence to optimize every step along the way. By layering AI agents on top of it, buyers can move beyond static bidding rules to dynamic strategies that learn, adapt, and act in real time. This not only delivers smarter campaigns and better outcomes but gives buyers a competitive edge in an ecosystem that is becoming increasingly automated.

How we’re enabling partners to build their own AI solutions

Unlike many offerings today, FreeWheel doesn’t want to dictate our clients and partners’ AI strategy (remember the ice cream?). Our focus is helping enable it – with our data, infrastructure, and tools to power agents and models that help deliver and obtain the best results. So, we’ve also made it possible for our clients and partners to build toppings on our platform, including building customized bidding agents directly into our Buyer Cloud.

And this isn’t a future state; this is now. Today, over 30 customers have built their own technology platforms, leveraging our tools like customizable bidding agents and data augmenter to drive better results. By leveraging FreeWheel’s Buyer Cloud API structure, one client was able to directly integrate with an optimization engine to deploy custom solutions and transparent, automated workflows that resulted in +40% improvement in performance efficiency.

Next year, we’re going to continue to build out our APIs and our ecosystem to make it easier for clients and partners to build on agentic solutions, using all of the advanced tools available within the platform

Powering the future of streaming advertising

AI will continue to drive innovation and evolution across the entire industry. FreeWheel has AI embedded into our blueprint, giving buyers and sellers the choice, transparency, and scalability to unlock the power of AI in ways that best suit their needs so they can thrive in a fragmented, complex streaming ecosystem.