Eventus DIGITALReadiness check
← Insights

PE Value Creation

The market has already marked down your company's valuation. Nobody has told you yet.

Accenture missed its forecast by less than half a percent and lost close to a fifth of its value in a day. If you are paid for time, seats or day rates, the same question is being asked about you. Five reasons operators give for staying put — three are excuses.

George Davies · 27 September 2026 · 10 min read

Accenture missed its revenue forecast by less than half a percent in June. Its shares fell about 18% in a day.

Not because of the miss. Because investors already doubt the future of being paid by the hour. Revenue came in at $18.7bn against a consensus of $18.78bn, and guidance was trimmed by a point at the top end. A miss like that would once have cost a stock a few percent. This one erased close to a fifth of the company's value in a day.

−18% in a day

Accenture's share price after missing forecast by less than 0.5%.

If your business is paid for time, seats, licences or day rates, the same question is now being asked about you. When buyers value your business, they compare it with listed firms like Accenture. Those firms are being marked down because AI makes the work they charge for faster and cheaper to do. If you are paid the same way, your value is being marked down with them.

Staying put feels safe. It is now the riskier choice.

Level 2 is a contract priced on inputs: the day rate, the seat, the licence, the ticket. Level 3 is the same service priced on outcomes. Operators give five reasons for staying at Level 2. Three are excuses. Two are challenges that need a deliberate approach, and even those are no reason to stand still.

Reason one: our client is not asking for it

Paul Chapman, commenting on last week's piece, had spoken to commercial leaders across a number of IT services businesses. Most agreed outcome-led pricing is the future. They differed on pace, much of it down to a sense that the client's own appetite is not there yet.

TSIA's research, published by Luke Ronkowski, found that cost-plus and market-based pricing is the only pricing family in its study tied to negative growth. McKinsey's Michael Birshan said in November 2025 that about a quarter of the firm's global fees now come from outcome-based pricing — inside a business built on the day rate. In regulated sectors, the FCA's Consumer Duty already requires firms to show client outcomes.

Structurally, a Level 2 supplier has almost no incentive to raise the question. If the fix is fewer heads or fewer billable hours, saying so costs the supplier revenue. Silence from the client is frequently not market evidence. It is evidence nobody asked.

A 2023 study of 437 professional B2B purchasers found it is not risk that drives resistance to gain-share. It is fairness. Offered a 50/50 split with no fixed fee, they were markedly less willing to switch. Offered a 75/25 split in their favour plus a fixed fee, they were more willing, even where the 50/50 deal was worth more on paper.

Verdict: an excuse. Your client is not asking because you have never asked them. And when you do, offer a split that feels fair.

Reason two: we do not know how to write the contract

This looks like a legal drafting problem. It rarely is. A 2024 Simon-Kucher study named five real barriers: poorly defined deliverables, not enough cost or willingness-to-pay data, cost that varies across delivery, pricing decided partner by partner, and procurement built for the old model. None sit in a contract template. All are capability gaps.

The deeper problem is that the contract gives the supplier no way to be paid for making the client more efficient. Fewer hours, fewer seats, less revenue. And with no plan for improvement, there is no agreement on where the client is starting from. That is the Attribution Gap: the gap between what a supplier delivers and what it can prove it delivered.

The fix already exists. The contract signs in two stages. Stage one pays for measuring where the client is starting from, with every term fixed in advance: length, the cap and floor on fees, the rates and the definitions. Stage two signs off that starting point, and the outcome fees start.

Jana Zimova, a strategic CFO, asked who defines success, who verifies delivery, and who carries the risk when the client does not do its part. The two-stage contract answers all three: success is defined from a baseline both sides sign, delivery is checked on one data set both can see, and the client's obligations are written into stage one.

Verdict: an excuse. The contract mechanisms exist. What is missing is a model that pays the supplier for making the client more efficient, and an agreed starting point to measure it from.

Reason three: headcount pricing is not going away that fast

The Accenture fall was not an isolated wobble. EPAM fell about 9% the same day, and is down close to two thirds from its January high. Cognizant fell to a 52-week low and was downgraded on AI risk. IBM fell around 5%. Four companies, one month, one shared fear: seat compression — AI agents doing the work that used to need a licensed seat per person.

Software now trades on a lower forward earnings multiple than the S&P 500. HFS Research and IBM found only 16% of senior executives expect to still be on headcount pricing within two years, down from 49% today. KPMG and HFS forecast labour-led delivery falling from 55% to 37% of the market.

There is a complication. Gartner found only 20% of customer service leaders had cut staff because of AI as of October 2025, and predicts half will rehire by 2027. But Gartner reports what has happened so far; the stock moves price what happens next. Repricing runs ahead of the layoffs.

Rather than referee a forecasting argument, check your own accounts for five signs that show up earlier than any report:

  1. 1

    Unprompted automation questions

    The client asks whether a task could be automated, before you have raised outcome-led pricing.

  2. 2

    AI benchmarking

    Procurement benchmarks your day rate against an AI subscription, not a competitor's day rate.

  3. 3

    Scope cuts at renewal

    A renewal opens with fewer seats or hours, rather than a lower price.

  4. 4

    Unbilled AI gains

    Your own team uses AI to work faster, and nobody has adjusted what is billed.

  5. 5

    New faces on calls

    The client's automation or AI team starts joining calls that used to be purely operational.

Verdict: an excuse. The market repriced this in June. Two or more of the five signs on the same account, and it has already started there.

Reason four: it disrupts a working relationship

Real, and worth admitting. Negotiating an outcome-led contract takes longer than signing a day-rate renewal. There is no reliable published figure for how much longer, so none is offered here.

Verdict: real. It costs time. That is not a reason never to start. Start on an account that can absorb the disruption.

Reason five: we do not have the capacity to build this yet

Also real, and it scales with size. M&A advisers generally cite recurring, outcome-priced revenue in business services at 6 to 8 times EBITDA, and project and time-and-materials work at 3.5 to 5 times. Segments with more recurring revenue trade at close to double the multiple of mostly project and resale segments.

6–8× vs 3.5–5×

EBITDA multiples for outcome-priced recurring revenue versus project and time-and-materials work.

Greg Head made the buy-side point: renewal history on outcome-priced contracts is usually what lets a buyer underwrite the higher multiple. Every renewal during the hold adds evidence to the exit story. For a small operator with one or two accounts, that fixed cost is not yet repaid — waiting, or starting narrow, is rational there.

Verdict: real, on small accounts only. Above that, waiting is the expensive choice.

The verdict

Three of the five reasons are excuses: client appetite, contract mechanics and market pace. Two are challenges: the cost of disruption, and capacity on accounts too small to justify it yet. All three excuses hide the same problem. You cannot prove what your work is worth to the client.

Changing the price is the last step, not the first. The commercial model decides how you are paid. The operating model decides whether you can deliver the outcome, and prove you did. Start with one account — one that can absorb the disruption, where you already hold some of the data. Not your hardest relationship. Not your smallest.

  1. 1

    Our client is not asking for it

    Excuse. Nobody has asked them. Offer a split that feels fair, not 50/50.

  2. 2

    We do not know how to write the contract

    Excuse. The mechanisms exist. Build a model that pays you for efficiency, from an agreed starting point.

  3. 3

    Headcount pricing is not going away that fast

    Excuse. The market repriced this in June. Check for the five early signs.

  4. 4

    It disrupts a working relationship

    Real. It costs time. Start on an account that can absorb it.

  5. 5

    We do not have the capacity yet

    Real, on small accounts only. Start on one account where you already hold data.

This is not a cycle. It is a market shift, and AI is accelerating it. In a shift, the cautious choice and the safe choice stop being the same thing. The question is not whether you can afford to move. It is whether you can afford not to.

Sources

  1. Paul Chapman, Jana Zimova and Greg Head, public comments on the 20 September 2026 edition, credited by name.
  2. TSIA, via Luke Ronkowski, Senior Director of Managed Services Research: cost-plus and market-based pricing tied to negative growth.
  3. McKinsey, Michael Birshan, quoted in Business Insider, 17 November 2025.
  4. FCA Consumer Duty, in force from 2023.
  5. Keränen, Kienzler, Salonen, Terho and Totzek, Industrial Marketing Management, November 2023: 437 professional B2B purchasers.
  6. Simon-Kucher, June 2024: five barriers to value-based pricing in professional services.
  7. TheStreet, February 2026, from Goldman Sachs Research (Matthew Martino): software forward P/E below the S&P 500.
  8. IT services stock moves, 18 June 2026 (Accenture, EPAM, Cognizant, IBM): The Motley Fool, TIKR, TradingKey and Investing.com.
  9. HFS Research and IBM; KPMG and HFS Research; Jefferies, February 2026.
  10. Gartner, 3 February 2026: 20% of customer service leaders had cut staff due to AI; half predicted to rehire by 2027.
  11. EBITDA multiple spread: general M&A advisory commentary, indicative ranges; Aventis Advisors IT services valuation index, 2015–2026.