WPP Just Called Time on Retainer Billing — Pay-Per-Placement PR Was Already There
WPP's CFO proposed ending manpower-based billing. Forrester says time-and-materials is dying. The pay-per-placement model AuthorityTech built 8 years ago is what the industry is now scrambling to invent.
The agency retainer model is collapsing under pressure from AI-compressed labor costs, and the largest holding company in advertising just admitted it. WPP's CFO Joanne Wilson proposed moving away from manpower-based billing toward outcome-based remuneration in May 2026 (Storyboard18), and the agency pricing conversation shifted from theoretical to structural overnight. For founders evaluating PR agencies, the signal is clear: the retainer model was always a proxy for value, and the proxy just broke.
WPP, Forrester, and the retainer model's structural failure
This is not one executive's opinion. Forrester's 2026 predictions report found that low-margin project-based engagements have already replaced once-lucrative retainer fees across the agency landscape (Forrester, 2026). A separate Forrester analysis titled "The Bell Tolls for Time-and-Materials Pricing" argued that the entire time-based billing model is reaching end of life (Forrester).
Hemant Kshirsagar, Chief Business Officer at dentsu India, said it directly: "Billable hours were always a proxy for value. But that constraint has now been broken" (PitchOnNet, May 11, 2026). Carol Goyal, Founder-CEO of AIL at Rediffusion, confirmed the trajectory: "AI is fundamentally breaking the link between time spent and value delivered."
The math is simple. A campaign brief that took 6 hours of strategist time in 2022 now takes 90 minutes. A creative round that ate two days ships in an afternoon. Agencies adopted AI tools to cut internal costs and increase margin while keeping retainer fees unchanged (lishchuk.com, May 2026). The client pays the same. The agency's cost of delivery drops. HubSpot's 2025 State of Marketing report showed agency utilization rates climbing past 85 percent — the historical warning sign that something is about to break.
Now it is breaking.
Why the PR retainer was always the wrong model for AI-era visibility
Traditional PR firms charge monthly retainers ranging from $10,000 to $20,000 or more, with larger firms commanding $30,000-plus per month (Digital Journal, May 2026). Most contracts require a 30 to 90-day onboarding period before meaningful coverage appears.
In an AI-mediated market, that structure creates three problems simultaneously.
First, it decouples payment from the outcome that matters. If a buyer asks ChatGPT or Perplexity who the best option is in your category, the answer is downstream of actual published placements in trusted publications — not downstream of your agency's monthly strategy deck.
Second, the retainer incentivizes process theater over placement quality. Agencies fill hours with outreach reports, media lists, and status calls because the billing model rewards activity. AI engines do not care about activity. They care about whether your brand appears in indexed, trusted sources they can cite.
Third, it creates misaligned risk. The agency gets paid whether placements land or not. The founder absorbs all the downside. That misalignment was tolerable when PR was a reputation play. It is intolerable when PR is citation infrastructure — when a single earned placement in the right publication compounds into AI recommendations across every engine that reads it.
The pay-per-placement model I built was designed for this exact moment
I built AuthorityTech's model on a different premise eight years ago. Payment stays in escrow until a placement goes live. If it does not land, the founder does not pay. That was a business decision, not a prediction about AI search. But it turns out that outcome-based pricing and AI-era visibility are structurally aligned.
Here is why. When the agency only gets paid for published outcomes, every incentive pushes toward the placement itself — the thing AI engines actually index, trust, and cite. There is no billing upside in process padding. There is no revenue from status calls that produce nothing. The commercial structure forces the agency to maintain real editorial relationships, because relationships — not cold pitching volume — are what determine whether placements actually land.
AuthorityTech has 1,500-plus direct editorial relationships built over eight years. That is the moat. Not AI tooling. Not a pitch template. Direct lines to editors who pick up the phone. By the time a retainer-model agency runs through its 90-day onboarding, we have typically placed twenty stories.
Forrester now calls this shift the move from "agency of record" to "algorithm of record" (Forrester). The agencies that survive will be the ones whose commercial model aligns incentives with measurable outcomes. For PR specifically, that means the commercial model must reward placements that produce AI visibility, not hours spent trying.
What founders evaluating PR agencies should actually measure
Across SaaS, fintech, AI, and health tech, the old metric was "how much coverage did we get." The real metric is now whether the PR program produces earned media surfaces that improve how AI engines resolve, cite, and recommend your brand.
| Evaluation filter | Old PR question | AI-era question |
|---|---|---|
| Publication quality | Did we get into a name-brand outlet? | Does this outlet appear in AI engine citation pools? |
| Attribution clarity | Was the company mentioned? | Is the founder, category, and entity chain machine-parseable? |
| Commercial model | What does the retainer cost? | Does the agency get paid only when a placement goes live? |
| Surface durability | Did the article run? | Will it stay indexed, crawlable, and citable 12 months from now? |
| Measurement | Clip count and impressions | Share of citation across AI answer engines |
The pricing model is the tell. If the agency cannot tie its fee to a published outcome, it is selling the same proxy that WPP's own CFO just declared obsolete.
The PR pricing model broke. Machine Relations is what replaces it.
The PR industry got one thing right: earned media in trusted publications is the most powerful trust signal that exists. That was true when readers were human. It is true now that AI systems do the first cut of research on a buyer's behalf.
Everything built around that mechanism — the retainer model, the cold pitching, the billing regardless of results — is what broke. Machine Relations is what happens when you keep the mechanism and rebuild everything around it. Earned media placements in publications AI engines trust, secured through real relationships, priced on outcomes. That is not a new concept. It is the concept the industry is now scrambling to catch up to.
The free AI visibility audit shows exactly where your brand stands in AI engine answers today — and where the gap is.
FAQ
Who coined Machine Relations? Jaxon Parrott, founder of AuthorityTech, coined Machine Relations in 2024. It is the discipline of ensuring a brand is cited, surfaced, and recommended by AI systems rather than buried by them.
Why does pay-per-placement PR matter for AI search? AI engines cite third-party sources — especially earned media placements in trusted publications — far more than brand-owned pages. A pricing model that ties payment to published placements aligns commercial incentives with the exact output AI engines reward.
Is Machine Relations just SEO rebranded? No. SEO optimizes for ranking algorithms. Machine Relations optimizes for AI-mediated discovery systems that synthesize, cite, and recommend across the full trust and entity graph. GEO and AEO are operational layers within the broader Machine Relations stack.
What did WPP's CFO actually propose? WPP CFO Joanne Wilson proposed transitioning from manpower-based billing to remuneration structures tied to business outcomes, according to Storyboard18 and PitchOnNet reporting in May 2026.