Stop Delivering. Start Proving.

The technology industry is destroying the pricing models that sustained it for six decades. Artificial intelligence makes products more valuable while simultaneously eliminating the users those products are priced around. The resolution lies in a capability most service businesses do not yet possess: the ability to connect what they deliver to what their clients achieve.

Most managed services businesses operate at Level 2, a model built entirely around inputs. The commercial prize lies at Level 3 and Level 4. Moving to Level 3 alone delivers measurable improvement in revenue retention, contract length, and enterprise value. Level 4 amplifies that further. Neither transition requires wholesale reinvention. Both require proof, and the distance between what a business delivers and what it can prove it caused, the Attribution Gap, is what stands in the way.

In thirty years of operational leadership, I have sat in a great many renewal conversations. The dynamic is almost always the same. The vendor has delivered. Service level agreements have been met. And yet the conversation is about price. Not about value. Not about outcomes. About price.

For six decades, technology businesses priced on inputs, per seat, per user, per device. Artificial intelligence has exposed the flaw in that model. It automates tasks, eliminates users, and destroys the very units on which revenue was built. The product becomes more capable. The licence count falls. Revenue contracts even as value delivered expands. The Technology Services Industry Association calls this the value paradox: AI makes products exponentially more valuable while simultaneously vaporising the revenue model that made them profitable.[1][2]

Traditional SaaS businesses are not immune. Atlassian reported its first-ever decline in enterprise seat counts in early 2026. Workday announced layoffs driven by AI efficiency, eliminating the very roles its software prices around. The SaaSpocalypse of February 2026 wiped 285 billion dollars from software market capitalisations in a single day, not because AI failed, but precisely because it succeeded. AI-native startups face the same problem from a different angle: inference carries real compute cost, so revenue and cost scale together. The startup model is not a disruption of the incumbent model. It is a different version of the same structural problem.[3]

Artificial intelligence does not eliminate services. It makes services mandatory. The last mile of enterprise AI adoption demands a powerful services layer to bridge the gap between raw capability and real-world outcomes.

The Technology Services Industry Association’s B4B framework, developed by Thomas Lah and J.B. Wood, defines four levels of supplier maturity. Most businesses operate at Level 2: input-based commercial models where the supplier prices on effort, headcount, licences, and time. The client owns all outcome risk. At renewal, years of delivery count for very little when the only differentiator the buyer can see is price. Net revenue retention sits at 90 to 100 per cent. The switching cost is contractual, not commercial.

Level 3 is where the supplier begins managing the client’s progression toward a result. Adoption is monitored. Health scores are tracked. Consumption gaps are closed. Progress is reported, not just activity. The commercial impact is quantifiable: NRR moves toward 110 per cent and above, contracts extend, and the switching cost shifts from contractual to relational. Research shows NRR above 110 per cent can add one to two turns to the valuation multiple. Level 3 is not a stepping stone. For many businesses it is the right commercial destination, and the returns are material.[5][6]

Level 4 is outcome-based contracting. The client pays for what is achieved. The switching cost becomes evidential: the client holds a body of proof linking the supplier’s service to their business performance, and walking away means losing it. The margin expansion is larger, and the multiple re-rating is greater than at Level 3. But the prize is available at Level 3 too. The worst commercial outcome is not failing to reach Level 4. It is staying at Level 2.

he Attribution Gap is the distance between what a service business delivers and what it can prove it caused. Most service businesses live permanently in this gap. They deliver services, hit service level agreements, complete projects on time, and then, at renewal, cannot answer the one question that determines whether they retain the contract on their terms or capitulate on price.

Which specific business outcomes did your service drive, and do you have the evidence to prove it?

Without an answer, every renewal is a price negotiation. Every tender is a comparison exercise. Every portfolio review is a defence. You have built value you cannot monetise.

The gap exists because outcome contracts attempted without rigorous attribution fail. Disputes are predictable: seasonality, marketing activity, staff changes, competitor actions. Without a methodology separating what the service caused from what would have happened anyway, every claim is contestable. Modern attribution modelling makes it possible to establish with statistical rigour which actions caused which outcomes. The Attribution Gap is not a data problem. It is a methodology problem. And it is solvable.

A managed services business that transitioned from Value in Exchange to Value in Use saw annual gross profit on a single 300-seat client move from £75,600 to £184,350, a 144 per cent increase, with the same number of seats and no new technology. The only thing that changed was proof.

Over three years, total contract value moved from £756,000 to £1,154,000. Gross margin expanded from 30 per cent to 46 per cent. Contract length moved from rolling annual to a three-year commitment. The client paid more not because the service improved, but because the value was visible, evidenced, and commercially undeniable. This was achieved by restructuring the commercial model around three components: a base service fee, an efficiency capture mechanism sharing verified automation savings, and a performance bonus tied to agreed outcome thresholds. Each was measurable. Each was defensible. None required the client to take anything on trust.[4]

The commercial improvements from moving up the maturity ladder do not only improve annual performance. They restructure the metrics that determine what a business is worth. Enterprise value in technology services is driven by gross margin, net revenue retention, and contract quality. Both Level 3 and Level 4 move all three in the right direction.

At Level 3, NRR moves toward 110 per cent and above. Research is consistent: NRR above 110 per cent can add one to two turns to the valuation multiple. For a business at five million pounds of recurring revenue, that is five to ten million pounds of additional exit value from a change in operating model alone. Level 3 also improves margin through the monetisation of adoption and optimisation services that most businesses currently give away.[6][10]

At Level 4, the impact compounds. Gross margin expands because outcome-based revenue is earned on already-resourced contracts with no incremental delivery cost. Businesses with NRR above 120 per cent command a 30 to 50 per cent higher valuation multiple than comparable businesses at 100 per cent. A ten-point NRR improvement can add 20 to 30 per cent to enterprise value. Converting a meaningful portion of the client base from Level 2 to Level 3 or Level 4 does not only improve current-year EBITDA. Its re-rates the multiple at which that EBITDA is valued.[10][11]

The businesses that will achieve the strongest exits in the next three to five years are not those with the best technology. They are those that can demonstrate, through evidence, that their clients are growing in value, and that the growth is attributable to them.

In cyber services, IBM’s 2025 Cost of Data Breach report puts the average breach cost at $4.44 million, with a mean time to identify and contain of 241 days. A client paying for monitoring activity has every right to ask what it prevented. The commercial model that survives is not the one that counts vulnerabilities patched. It is the one that evidences the financial value of the risk eliminated.[7]

In infrastructure services, uptime is table stakes. The market is moving from Service Level Agreements to Experience Level Agreements, measuring whether infrastructure performance delivered measurable improvement in business outcomes. Providers still pricing on device counts and response times are not competing on the right axis.[8]

In financial services, the transition from activity to outcomes is no longer voluntary. The FCA’s Consumer Duty has made evidenced client outcomes a regulatory requirement. The FCA expects boards to monitor outcomes, not processes. Vendors still selling effort face contracting pressure and commoditisation risk. The firm that can demonstrate, with data, which specific client outcomes its service caused is not just differentiated. In a Consumer Duty environment, it is protected.[9]

The prize is available at Level 3 and at Level 4. The size differs but the direction is the same: more revenue from existing clients, higher margin, longer contracts, and a better multiple at exit. Every business that remains at Level 2 is competing on price, losing margin to inertia, and leaving enterprise value on the table.

The winners will not be those with the best artificial intelligence. Everyone will have good artificial intelligence. The winners will be those who master the delivery, pricing, and evidencing of the business outcomes that artificial intelligence enables.

The Attribution Gap is the most important commercial problem in the technology services industry right now. It is solvable at Level 3 as well as Level 4. The question is not whether to move. It is whether your business moves before the market reprices around the businesses that already have.