Cuebiq’s Brennan Lake: Location Data Can Now Steer CTV Bidding In Real Time
Closed-loop advertising has long been a walled garden privilege. Google and Meta could connect impressions to outcomes because they owned identity, inventory, and measurement inside a single ecosystem. A new coalition approach is bringing that capability to the open web and CTV. And they’re doing it without any single party controlling all three.
“Walled gardens could kind of close the loop without any third party participation. Now that we have a coalition of partners that can bring closed loop advertising to the open web, we’re almost offering a more neutral value proposition because the entity selling the media and the entity grading it are independent from each other,” Brennan Lake, vp of partnerships at Cuebiq, told Beet.TV contributor David Kaplan at the Beet.TV Leadership Summit with Omnicom Media during Advertising Week New York.
The integration works through PurePlay AI, which sits inside an SSP and scores CTV impressions in milliseconds based on their propensity to drive store visits. Those scores pass bid signals to a client’s existing DSP, which bids up on high-value households, bids the floor on baseline impressions, and declines to bid on households that have already converted.
Results Shine from Sunglass Hut
A multi-year campaign with Luxottica’s Sunglass Hut demonstrates what the model delivers in practice. Running through PurePlay, the brand saw a 17% improvement in cost per store visit and an 8% improvement in incremental lift compared to publisher-direct buying.
“17% cheaper to get users in store through advertising,” Lake said.
Data quality raises the stakes for agentic models
When location data feeds a live bidding loop rather than a post-campaign report, the tolerance for bad data drops significantly. Lake cited one tell-tale quality check: some location data sets show visitors at Chick-fil-A on Sundays, when the chain is closed.
“If you’re feeding that into an agentic model for real-time bidding optimization, at best it might lead to subpar optimization, but at worst it could really misallocate your ad spend,” Lake said. “Having 100% real data is absolutely table stakes.”
Beyond data accuracy, longitudinal panel persistence matters. Observing opted-in users over months and years enables the model to capture competitive behavior, brand loyalty, and outcomes in ways that point-in-time panels cannot.
How automation frees up creative and strategy
As bidding optimization moves into agentic models, the human role shifts rather than disappears. Manual intervention in real-time bidding is what goes away. Creative, media planning, and strategic oversight remain squarely human responsibilities.
“That type of automation almost gives the luxury and more bandwidth to the human to actually focus on what humans do best like creative, like media planning,” Lake said. “What goes away is the need for manual intervention, not the oversight, not the creative, not the media planning.”
Low friction for brands
Many brands and agencies buying CTV already have DSP seats set up where PurePlay operates and have measured with Cuebiq previously. The ask is simply to route existing buys through the model and compare KPIs.
“You’ve been running publisher direct,” Lake said. “Try running it through PurePlay’s models, give you a no-cost trial, and let’s see how those KPIs move on things like cost per store visit and incrementality.”
You’re watching “The Beet.TV Leadership Summit with Omnicom Media at Advertising Week 2026,” presented by Cuebiq, Databricks, Newton Research, Roku and Teads. For more videos from this summit, please visit this page.
Closed-loop advertising has long been a walled garden privilege. Google and Meta could connect impressions to outcomes because they owned identity, inventory, and measurement inside a single ecosystem. A new coalition approach is bringing that capability to the open web and CTV. And they’re doing it without any single party controlling all three.
“Walled gardens could kind of close the loop without any third party participation. Now that we have a coalition of partners that can bring closed loop advertising to the open web, we’re almost offering a more neutral value proposition because the entity selling the media and the entity grading it are independent from each other,” Brennan Lake, vp of partnerships at Cuebiq, told Beet.TV contributor David Kaplan at the Beet.TV Leadership Summit with Omnicom Media during Advertising Week New York.
The integration works through PurePlay AI, which sits inside an SSP and scores CTV impressions in milliseconds based on their propensity to drive store visits. Those scores pass bid signals to a client’s existing DSP, which bids up on high-value households, bids the floor on baseline impressions, and declines to bid on households that have already converted.
Results Shine from Sunglass Hut
A multi-year campaign with Luxottica’s Sunglass Hut demonstrates what the model delivers in practice. Running through PurePlay, the brand saw a 17% improvement in cost per store visit and an 8% improvement in incremental lift compared to publisher-direct buying.
“17% cheaper to get users in store through advertising,” Lake said.
Data quality raises the stakes for agentic models
When location data feeds a live bidding loop rather than a post-campaign report, the tolerance for bad data drops significantly. Lake cited one tell-tale quality check: some location data sets show visitors at Chick-fil-A on Sundays, when the chain is closed.
“If you’re feeding that into an agentic model for real-time bidding optimization, at best it might lead to subpar optimization, but at worst it could really misallocate your ad spend,” Lake said. “Having 100% real data is absolutely table stakes.”
Beyond data accuracy, longitudinal panel persistence matters. Observing opted-in users over months and years enables the model to capture competitive behavior, brand loyalty, and outcomes in ways that point-in-time panels cannot.
How automation frees up creative and strategy
As bidding optimization moves into agentic models, the human role shifts rather than disappears. Manual intervention in real-time bidding is what goes away. Creative, media planning, and strategic oversight remain squarely human responsibilities.
“That type of automation almost gives the luxury and more bandwidth to the human to actually focus on what humans do best like creative, like media planning,” Lake said. “What goes away is the need for manual intervention, not the oversight, not the creative, not the media planning.”
Low friction for brands
Many brands and agencies buying CTV already have DSP seats set up where PurePlay operates and have measured with Cuebiq previously. The ask is simply to route existing buys through the model and compare KPIs.
“You’ve been running publisher direct,” Lake said. “Try running it through PurePlay’s models, give you a no-cost trial, and let’s see how those KPIs move on things like cost per store visit and incrementality.”
You’re watching “The Beet.TV Leadership Summit with Omnicom Media at Advertising Week 2026,” presented by Cuebiq, Databricks, Newton Research, Roku and Teads. For more videos from this summit, please visit this page.
Series Videos
