WPP Launches Open Pro AI Platform, Effectively Bypassing Its Own Agencies

WPP just launched Open Pro, a self-serve AI platform that takes a brand from creative brief to programmatic buy without touching an agency. Google was among the pilot users. A holding company that runs on agency fees shipped a tool designed to make agencies optional. Whether that is genuine self-disruption or a cheaper front door to the same house is the right question to ask.

The margin problem

Strip the disruption language and this is a margin story. A 15% commission on a $500K media buy is high-margin because the marginal cost of the next dollar is near zero. A subscription carries real product, hosting, and support cost. If Open Pro scales, WPP's revenue may hold while gross margin quietly compresses. The read-across hits Publicis, Omnicom, and IPG the moment investors decide the category is repricing from fee to SaaS. WPP is trading a fat, lazy revenue line for a leaner, harder-working one and calling it innovation.

Why the volume may never come

For this to matter commercially, mid-market CMOs at $10M to $50M in spend have to want to own media strategy risk, brand-safety calls, and creative compliance with no agency to blame when it goes wrong. Most don't. And the plumbing matters: Open Pro routes through WPP's own media tooling, so WPP still clips margin, just further upstream. Google piloting it proves nothing about demand. Google wants every holdco dependent on its pipes and is happy to lend a logo to a press release that deepens that dependence.

The internal damage is real regardless

Here is where the "just a press release" read falls short. The tool doesn't have to win in the market to blow up WPP's internal org chart. Once Open Pro routes around the agency layer, GroupM's internal transfer pricing stops making sense, and WPP's own CFO starts asking what the desk actually adds. That interrogation happens within a quarter or two of the first real volume, not years out. The junior planners and traffickers whose margins were already thin get cut first. Open Pro can be a commercial dud and still force a painful internal repricing.

Our call

Through WPP's next two earnings calls (FY2025 results in early 2026 and H1 2026), WPP will not disclose Open Pro revenue, ARR, or active-client counts as a standalone figure. It will stay folded into WPP Media commentary. Confidence: medium.

The launch shipped with no disclosed revenue, no client commitments beyond a Google pilot, and no evidence of real programmatic volume moving through the pipe. Holdcos only carve out a new line when the number helps the story. A small, margin-dilutive subscription base set against 15% commissions helps nobody in an investor deck. If WPP breaks out Open Pro as its own disclosed line with hard adoption metrics before August 2026, we're wrong.

WPP built a self-serve AI platform, Open Pro, that lets a brand go from creative brief to programmatic buy without ever touching a full-service agency. The strategic knot: a holding company that lives on agency fees just shipped a product designed to make agencies optional. Google was among the pilot users. That's the news. The question for every holdco and ad-tech operator watching is whether this is genuine self-cannibalization or a cheap lead-gen funnel that borrows disruption language to dress up what is really a new front door to the same house.

Reversibility: Type 1 for WPP's own P&L. Once you tell the market the durable asset is the pipe and not the people, you can't un-say it, and you can't easily rehire the trading-desk layer you cut. Type 2 for everyone else, who can watch and copy.

What's actually being decided: not "should brands self-serve" but "where does margin live in a holdco once AI does the orchestration." The forcing function is Q4 2025 procurement pilots and the next few holdco earnings calls, where analysts will ask what SaaS revenue does to blended margin.

The Market Analyst. Strip the disruption language and this is a margin story, not a revenue story. A 15% commission on a $500K media buy is high-margin because the marginal cost of the next dollar is near zero. A subscription that replaces it carries real product, hosting, and support cost. If Open Pro scales, WPP's revenue may hold while gross margin quietly deteriorates, and the read-across hits Publicis, Omnicom, and IPG the moment investors decide the whole category is repricing from fee to SaaS. In plain terms: WPP is trading a fat, lazy revenue line for a leaner, harder-working one and calling it innovation. The contrarian trade is short holdcos on margin, not on top line.

The Skeptic. What has to be true for this to matter? Mid-market CMOs at $10M to $50M in spend have to want to own media strategy risk, brand-safety calls, and creative compliance with no agency to blame when it goes wrong. Most don't. And read the plumbing: Open Pro routes through WPP's own media tooling, so WPP still clips margin, just further upstream. Google piloting it proves nothing about demand. Google wants every holdco dependent on its pipes and is happy to lend a logo to a press release that deepens that dependence. In plain terms: this looks less like WPP disrupting itself and more like WPP building a cheaper front door to the same house.

The Operator. Tuesday morning, the first thing that breaks isn't strategy, it's account management. When a monthly subscription replaces a commission on a real buy, the junior planners and traffickers whose margins were already thin get cut first. The nastier second-order effect lands inside WPP. If Open Pro routes around the agency layer, GroupM's internal transfer pricing stops making sense, and WPP's own CFO starts asking what the desk actually adds. That interrogation happens within a quarter or two of the first real volume, not years out. The tool doesn't have to win in the market to blow up the internal org chart. It just has to exist and get demoed to the board.

The Customer / End User. The mid-market brand is the whole ballgame, and nobody has shown it wants this. A brand spending $30M doesn't lie awake wishing it could fire its agency and personally own the programmatic dashboard. It wants fewer 2am brand-safety fires and someone accountable when the campaign misses. Open Pro hands them the wheel and, quietly, the liability. Some scrappy DTC brands will love it. Plenty will pilot it, hit the first compliance question they can't answer, and call an agency. In plain terms: self-serve sounds like freedom until you're the one who has to explain to the CEO why the ad ran next to something toxic.

Where the council splits. Three live disagreements. First, the Market Analyst and the Skeptic disagree on whether this is even a real business: one models margin compression as if volume comes, the other doubts the volume shows up at all. Second, the Operator says the damage is internal and near-term regardless of market traction, which cuts against the Skeptic's "it's just a press release" read. The tool can be a commercial dud and still force WPP's CFO to reprice GroupM. Third, the Customer's take undercuts the whole disruption narrative: if mid-market brands won't own the risk, Open Pro converges on the Skeptic's lead-gen funnel, and the SaaS-multiple dream the strategists love never arrives.

What it hinges on. Two beliefs. One, whether mid-market brands actually want to own media and brand-safety risk without an agency. Two, whether self-serve SaaS revenue can be repriced high enough to defend holdco margins. Both are unproven, and both point the same way if they fail: no disruption, but real internal cannibalization at the trading-desk layer. The council leans toward "smaller than the headline, more disruptive internally than externally." Before anyone at a rival holdco panics and ships their own portal, verify one thing: is real programmatic volume flowing through Open Pro, or is the pipe empty and the Google logo doing the PR? No disclosed ARR, no client commitments beyond a pilot. That silence is the answer for now.

Prediction: Through WPP's next two earnings calls (FY2025 results in early 2026 and H1 2026), WPP will not disclose Open Pro revenue, ARR, or active-client counts as a standalone figure, keeping it folded into WPP Media commentary.

Confidence: Medium. Holdcos hide new-product traction inside segment revenue until it's flattering, and there's nothing flattering to disclose yet.

Why: The launch shipped with no disclosed revenue, no client commitments beyond a Google pilot, and no evidence of programmatic volume moving through the pipe, which is exactly the profile of a product not yet worth breaking out. Holdcos only carve out a new line when the number helps the story, and a subscription revenue base that's small and margin-dilutive against 15% commissions helps nobody in an investor deck. The opposite outcome, a proud standalone Open Pro disclosure, would require both real adoption and management's willingness to show margin math that works against them, and neither is in evidence today.

Revisit by 2026-08-15: We're right if WPP's FY2025 and H1 2026 reporting keeps Open Pro inside aggregate WPP Media figures with no standalone revenue, ARR, or client-count number. We're wrong if WPP breaks out Open Pro as its own disclosed line with hard adoption metrics.