In recent days, multiple sources said any such council will bring together representatives from media-buying agencies, demand-side platforms, publishers, and supply-side platforms.

In a statement shared with Digiday, IAB Tech Lab CEO Anthony Katsur, said, “We are embarking on this journey with a first meeting that could evolve into a formal council that regularly convenes… our immediate priority is to ensure that the ultimate parties to a transaction, agencies and publishers, are brought together in one forum to define how they wish to conduct programmatic business.”

He went on to add, “The role of technical protocols and standards, such as those governed by the IAB Tech Lab, is to enable and operationalize the business frameworks agreed upon by these parties. However, they must be built on a clear, shared understanding of how the market wants to operate.”

The move follows months of tension sparked by one seemingly technical change in the open-source plumbing that powers much of the web’s advertising economy.

How we got here

In August, Prebid — the nonprofit consortium that maintains open-source programmatic infrastructure, primarily for sell-side players — introduced its 10.9 update, changing how transaction-IDs, or TID, are handled.

Previously, all SSPs shared one global identifier per impression — a manner that has been referred to as global-ID, or GPID. The update assigned a unique ID to each bidder, a change meant to curb data leakage and give publishers greater control over how much information buyers could piece together across auctions.

However, buy-side players pushed back, arguing the move hampered their ability to detect duplicate traffic and understand auction density, resulting in reduced transparency at the expense of efficiency. Sellers countered that the previous system allowed buyers to reverse-engineer inventory data and suppress yield.

That argument, long simmering, soon spilled into public view — notably, IAB Tech Lab maintained it was “materially non-compliant with OpenRTB” — with representatives of The Trade Desk, which subsequently launched OpenAds, emerging as leading voices in the matter.

‘Grand bargain’ proposals

As Digiday revealed last week, a cluster of such parties met in the days after this year’s October Prebid Summit during a closed-door meeting in New York City to test whether a middle ground could be formalized.

According to multiple attendees, participants floated a three-point “grand bargain” intended to restore equilibrium between the buy- and sell-sides:

• One request per impression: Publishers would send a single bid request for each ad opportunity, tagged with a TID, eliminating traffic duplication.
• Five bids per request: In exchange, DSPs would submit multiple bids — up to five per auction — to maintain competition and reveal a truer picture of market value
• Simplified floor pricing: Publishers would remove brand-level floor rules that complicate auctions and obscure pricing signals

Additionally, proponents of such a fix are understood to advocate that sell-side players adopt the recently proposed auction trading standard proposed by The Media Ratings Council.

While the proposals were exploratory, attendees described them as a first step toward a more disciplined supply chain — one where “doing the right thing” wouldn’t automatically cost publishers revenue.

However, there was some disagreement, with separate sources informing Digiday that there were divergent opinions, even among players with aligned interests, with a notable discrepancy in the reactions of different DSP representatives on hand. 

Buy-side vs. sell-side

The dispute over GPIDs and TIDs has re-exposed one of digital media’s oldest tensions: balancing the buy- and sell-sides’ interests while meeting privacy requirements. 

At the Prebid Summit this fall, The Trade Desk CEO Jeff Green warned that inefficiencies and opacity continue to sap confidence in the open internet. “Without improvement to the supply chain, we don’t see how the open web can compete with walled gardens,” he said during an on-stage exchange with Garrett McGrath, a Prebid board member and Magnite’s svp of product.

McGrath defended the update as a practical compromise. DSPs, he noted, had demanded that publishers pass TIDs in the first place. Prebid’s August change merely allowed them to do so without exposing excessive data. “It wasn’t about breaking visibility,” he later told Digiday. “It was about giving publishers optionality.”

For many advertisers and agencies, the fight remains largely invisible. Yet its outcome could shape how much of the open internet’s inventory remains competitive with the walled gardens of Google, Meta, and Amazon.

What we’ve heard

“Makes sense, they would say that, but can’t imagine they don’t have a top 100 client somewhere that spends less than 30% on Google.”

— One source reacts to Publicis Groupe’s recent claims that none of their top clients spends more than 10% of their media budget on large platforms.

Numbers to know

• $38 billion: The value of OpenAI and Amazon Web Services’ recent deal for access to Nvidia GPU access
• $102.35 billion: Alphabet’s latest quarterly earnings
• $17.7 billion: Amazon’s advertising revenue, per its latest quarterly earnings

What we’ve covered

• Ad tech’s economy depends on float — and it’s getting pricier to keep it moving

Nearly six in 10 invoices (58%) in the sector were paid late in the first half of the year, according to payment data from OAREX.

• News Corp-owned U.K. tabloid The Sun is building an AI agent for its programmatic business

British tabloid newspaper The Sun is building an AI agent for its programmatic advertising business, as it looks to prepare for when demand-side partners push agentic media buying at scale.

What we’re reading

• How Google fooled Brussels: The Tech Giant’s playbook exposed

Follow the Money explores Google’s efforts to offset the scrutiny of its market-leading ad exchange, popularly known as AdX.

• What OpenAI and other LLMs must do before building their ad businesses

Of all the predictions about AI, here’s one you can bet on with confidence: Major large language models will start building ad businesses in the very near future. We use OpenAI as an illustrative example, but the recommendations below apply to LLMs in general.

• The wrapper wars are: The Trade Desk strikes back at Prebid

See above.