Outcome-Based Pricing Is Real For Two Companies And A Rumor For Everyone Else

Accenture lost 18% of its market value in a single session this year. Worst day since 2016, for a company that's been public since 2001. The headline reason was a guidance cut and soft bookings. The real reason, the one analysts kept circling back to, was doubt about whether the firm can reprice its work fast enough to survive what AI is doing to the billable hour.

That's the tell. When a market that size moves that hard on a pricing question, the question is no longer theoretical.

Here's the pitch every consulting CEO is giving right now, in some version: we're moving away from time and materials, toward value. Pay for the outcome, not the hours. It sounds inevitable, and it probably is, eventually. But walk the floor of any of these firms and ask the delivery team the same question, and you get a different answer. I've watched this play out inside enough of these conversations to know the pattern: the CEO is enthusiastic, the delivery lead is skeptical, and finance is doing math nobody wants to see on a slide.

A COO said it plainly to a colleague of mine a few weeks ago, cutting straight across her own CEO's projection that a meaningful share of the firm's revenue would be outcome-based within two years: "We've never delivered anything purely on outcomes, and we couldn't do it tomorrow." That's not cynicism. That's someone who actually runs delivery telling you the truth about what it takes to price a result instead of a timesheet.

Most of what gets called outcome-based pricing right now is outcome language bolted onto a fixed-fee contract. The vendor takes on the downside risk of missing a KPI, keeps none of the upside if they beat it, and calls the whole thing transformation. That's worse than honest time and materials. At least T&M is priced for what it is.

Real value-based pricing requires three things most firms don't have yet. Delivery standardized enough that a result is repeatable across teams, not dependent on which senior engineer got staffed. Tooling that tracks whether the outcome happened without someone compiling a status deck by hand. And a contract that specifies what happens when the client is the reason the outcome didn't land, because client inaction is the number one cause of failed outcomes, not vendor underperformance. Build those three things first. The pricing model follows. Skip the foundation and you get a fixed-fee contract wearing a KPI as a costume.

Two public companies are actually doing the harder work, and it shows up as a cost, not a slogan.

Endava is deliberately restructuring its book toward outcome-based engagements, and management says so on earnings calls without softening it. Utilization drops during the transition, and that drop shows up in margin. Deals that used to close in two or three months now take three to four times as long, because client legal and procurement teams haven't built the muscle to negotiate this contract type yet. Endava's own numbers show its AI-driven business already running at higher margin than its traditional digital transformation book. That's what a real bet looks like. It costs something before it pays off.

Globant took a different path and built AI Pods: a token-metered subscription, roughly 100 million tokens for about $20,000 a month, with overage pricing and volume discounts for multi-year commitments. That's not outcome-based pricing in the strict sense. Nobody's getting paid for a verified business result. But it's a genuine break from the seat-and-hours model, because revenue stops being a function of headcount. Consumption pricing and outcome pricing solve different problems. Both are real departures from T&M. Neither is what most of the industry is currently selling.

Everyone else in this sector is somewhere between aspiration and risk disclosure. Accenture and Infosys talk about the shift constantly and disclose almost nothing concrete about what share of revenue is actually priced this way. EPAM's outcome-based story right now is analyst thesis about what its AI infrastructure bet could unlock, not a reported contract mix. And across the sector's regulatory filings, the language has shifted from silence to a defensive risk factor: clients are asking for outcome-based and transaction-based pricing, and getting it wrong could hurt margins. That's a real signal. It means the industry now treats this as a threat serious enough to disclose to shareholders. It is not evidence that anyone has solved it.

The market isn't waiting for the industry to sort this out either. Every name in this group traded down the same day, on the same macro noise, with no company-specific news driving any single stock. That's not proof the pricing question caused the move. It is proof that the market has priced in enough uncertainty about this entire sector's economics that a single geopolitical headline can knock all of them down together, including the two companies actually doing the harder work.

What that points to, for me, is a gap between how fast this transition gets talked about and how fast it can actually happen. AI compresses delivery time, which makes billing by the hour self-defeating for the vendor the moment automation gets good enough. That part is inevitable. What isn't inevitable is which firms get the operational foundation built before that pressure arrives, and which ones spend the next two years selling a KPI wrapped around a fixed fee and calling it transformation.

Sources: earnings calls and investor materials from Accenture, Endava, Globant, EPAM, Cognizant, and Infosys (2025-2026); Forbes Technology Council, "Fact Or Fiction: Is Outcome-Based Pricing For Services Firms A Reality?," July 16, 2026; market data July 22, 2026.