Observed Signal · Jul 31, 2026 · Business Model Change · Source: Digiday · Impact: 3/5 · Sentiment: Negative
Agencies Weigh Turning Compute Tokens into Media Bets
Advertising holding companies are exploring treating AI compute tokens like principal media inventory — buying tokens in bulk, pricing the risk, and reselling with a margin — to recoup rising AI costs that agencies have been absorbing on their balance sheets. The idea is divisive: a small Digiday poll of 58 respondents found 42% oppose agencies becoming token futures markets, fewer than 10% support it, and 36% are conditional on transparency. Proponents argue bundling token costs into principal media deals gives clients certainty and helps agencies fund infrastructure; critics warn of opacity, procurement benchmarking problems, conflicts of interest, and a failure mode where improved efficiency leaves agencies stuck with overcommitted token volume.
Potentially significant change to agency revenue models and client billing practices around AI compute costs; introduces new commercial and transparency risks but is not a platform-level policy change.
Track IAB Europe Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your free Explorer workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- A Digiday poll of 58 respondents found 42% said agencies shouldn’t become futures markets for tokens, fewer than 10% called it a legitimate revenue opportunity, and 36% said maybe if there is transparency.
- Some holding companies are already baking token (compute) costs into principal media deals and getting contracts signed on that basis.
- Many agencies have been carrying AI compute/token costs on their own balance sheets for nearly two years and are seeking ways to recover those expenses.
- Critics warn of opacity (clients not knowing what agencies paid for tokens or the margin applied) and a failure mode where agencies overcommit to token volumes and may need to sell at a loss or write them off if efficiency improves.
Connected Companies & Entities
5 Entities mapped“The IAB Europe’s chief economist Daniel Knapp put it plainly: “No one really knows quite yet how to price that in, so you’re seeing the busi...”
““What we saw in the last two years is that most agencies were carrying those costs entirely on their balance sheet, and they’ve been using i...”
“The AI companies have their own reasons to prefer it this way, too — servicing thousands of individual advertisers directly isn’t a business...”
“The AI companies have their own reasons to prefer it this way, too — servicing thousands of individual advertisers directly isn’t a business...”
““Principal trading is disclosed in holdco earnings as a growth line,” said Robert Webster, former WPP exec and founder of AI marketing consu...”
Ontology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
Ad industry wrestles with AI costs and value
As agencies move AI beyond pilots into daily workflows, computing and token usage costs are rising and forcing new governance and pricing models. Agencies and holding companies are experimenting with token caps, pooled access, subscriptions, and output-based fees, but struggle to measure the actual business impact of AI versus token and compute spend. Some firms (PMG, S4 Capital/Monks, Dept, Cheil, Publicis) are testing different approaches to manage or absorb AI costs while clients and procurement often expect lower fees from automation. The article highlights the unresolved measurement problem — tracking token spend is straightforward, attributing business value to that spend is not.
AI Compute Costs Reshaping Principal Media Deals
Agencies are increasingly routing AI infrastructure and token compute costs through principal media deals, requiring clients to commit fixed shares of spend to principal inventory in exchange for agencies covering AI expenses. The practice leverages agencies’ existing principal-media economics — bulk wholesale buys resold at a markup — to fund unpredictable AI costs, but it can create opacity for clients lacking audit rights. Industry voices say this is an extension of long-standing commercial trade-offs (efficiencies, offshoring, contract length) rather than a wholly new model, and outcome-based pricing remains rare outside well-aligned large brands.
Agencies’ AI Operating Systems Face Commoditization Risk
Agencies have invested in proprietary AI-powered operating systems (WPP Open, Omnicom’s Omni, Stagwell’s The Machine, Havas Converged, PMG’s Alli, Horizon’s Blu, etc.) to centralize data, planning and creative tools and increase client stickiness. A Gartner research note warned that roughly half of agency-built AI platforms could fail by 2029, citing consolidation, advertisers building bespoke AI or buying from hyperscalers, and shrinking differentiation. Agency and holding-company leaders (e.g., Omnicom’s John Wren, WPP’s Stephan Pretorius) argue scale and negotiated enterprise agreements give them an advantage, while vendors such as Brandtech’s Pencil emphasize transparency about underlying models (Anthropic, OpenAI) and token-costing. The ANA survey flagged principal-media buying is widespread but governance lags. The briefing frames a growing industry debate over AI token economics, vendor transparency and whether hyperscalers will disintermediate agency platforms.
Track Real-Time Market Signals & Shifts
Set up custom watchlists to receive automated, evidence-grounded executive digests whenever material signals or shifts occur across your tracked landscape.