Britain Made Buyers Publish a Verdict on Their Suppliers

← Back · ACT II · Chapter 11: Outcome Credibility Markets

Most of the governance arriving around technology purchases puts the duty on the seller. The vendor certifies, discloses, attests, and files, and the buyer receives the paperwork. The United Kingdom has done something different, and it took effect on 1 January 2026, for contracts awarded under the Act, with little notice outside procurement law. Under the Procurement Act 2023, a public buyer that has set key performance indicators for a contract must assess the supplier against them at least once every twelve months and on termination, and must publish that assessment. The buyer writes the verdict, puts its name on it, and anyone can read it.

That is the first instance I have found of a jurisdiction obliging the buying side to build and publish the performance record about its supplier. It is worth understanding properly, because it gets the architecture right and then stops one sentence short.

What the Act Actually Requires

Section 52 says that before entering into a public contract with an estimated value of more than five million pounds, a contracting authority must set at least three key performance indicators and must publish them. Section 52(6) puts frameworks, private utility contracts, concession contracts and light touch contracts outside that duty altogether, so a large share of public buying never reaches it at all. Section 71 says that at least once in every period of twelve months during the life-cycle of the contract, and on termination, the authority must assess performance against those indicators and publish information about that assessment. A separate duty in section 71(5) gives it thirty days to publish the circumstances where a supplier has breached the contract or failed to improve after a proper opportunity to do so. The assessment appears as a contract performance notice, rated good, approaching target, requires improvement, inadequate, or other.

Read as a buyer rather than a lawyer, that is serious machinery. The record is created by the party that paid, on a fixed cadence, in public, and it follows that supplier into every conversation a future buyer has about them.

The Sentence That Decides What Any of It Means

Section 52(4) defines a key performance indicator as “a factor or measure against which a supplier’s performance of a contract can be assessed during the life-cycle of the contract.” That is the entire definition. Nothing in it requires the indicator to describe the outcome the money was spent to produce. Delivery dates qualify. Ticket response times qualify. So does reporting cadence. Three measures of that kind, published in the contract details notice, satisfy the duty completely, and twelve months later the authority can rate the supplier good against all three without anyone having asked whether the result the contract was bought to achieve actually arrived.

There is already a public record of what departments choose to measure when they get to choose. The Cabinet Office publishes the indicators for central government’s most important contracts let under the old 2015 regulations, along with each supplier’s rating. The April to June 2026 file lists 4,689 indicators across 1,473 contracts in 17 departments. Of the ones given a performance rating, about nine in ten are Good. Hundreds of the indicators measure availability, response times or whether something happened on time. Not one of the 4,689 descriptions mentions the business case. The government is candid about what this adds up to: the page says the data “does not represent a comprehensive assessment of the performance of the service, the contract or the supplier.” A wall of Good ratings can sit beside a contract nobody has checked against the reason it was bought. For a supplier that delivers, that is a loss too, because its Good looks exactly like everyone else’s.

Section 52(2) then lets an authority set no indicators at all if it considers that the supplier’s performance could not appropriately be assessed by reference to them. That judgment is the buyer’s, and it is most likely to be reached on the contracts where value is hardest to pin down, which are the contracts where an honest assessment would be worth the most. The Act makes the record mandatory and leaves its subject to whoever is writing it.

There is a comparison worth making inside the same Act. Section 70 requires a contracting authority to publish specified information about any payment of more than thirty thousand pounds made under a public contract, before the end of thirty days beginning with the last day of the quarter in which the payment was made. The threshold is a number, the deadline is fixed, and what has to be published is not left to the authority to decide. When the subject is money leaving the account the drafting is exact. Section 52(4) came out of the same statute and produced a definition that almost anything satisfies. The looseness in the performance duty is not a limit on what legislation can specify. It is a decision about what the published record is for.

This is the gap I described in Your Contract’s Audit Right Can’t Tell You If the Purchase Worked, arriving through a different instrument. There the buyer holds a right reaching records and costs rather than the promised result, and here it holds a duty to publish a judgment whose subject nobody defined.

The Published Set Is Chosen When the Notice Is Written

There is a second decision inside this machinery, and it sits in the regulations rather than the Act. Regulation 39(4)(d) of the Procurement Regulations 2024 says the contract performance notice must carry “a description of the three key performance indicators set in accordance with section 52(1) which the contracting authority regards, at the time the contract performance notice is published, as most material to performance of the contract obligations.” Materiality is judged on the day the verdict is written, not on the day the contract was signed.

The Cabinet Office guidance on key performance indicators says the consequence out loud at paragraph 14: “These KPIs may be different from those set out in the contract details notice or in previous contract performance notices as the KPIs to be included must be those regarded as most material at the time the contract performance notice is published, so may change over time.” Paragraph 17 goes a step further. Where an authority considers that fewer than three of the indicators it set are still material, it “could (provided this would be a permitted modification under section 74 of the Act) consider introducing a new (or amending an existing) KPI to be assessed against and reported on”, and it should use the free text box in the description field to indicate which ones are new or amended.

The modification gate is real and I am not going to talk past it. Changing an indicator changes the contract, and section 74 has to be satisfied like any other change. What the drafting still allows is that the reported set moves between one notice and the next, with a free text box as the only trace a reader gets. Two notices about the same contract can be rated against different measures. Two suppliers in the same market can be rated against sets that were never comparable in the first place. None of that needs anyone to act in bad faith.

What the Same Government Specified in 2026

Section 70 is from 2023, and the comparison above rests on it. Two things written this year make the same point with less room to argue.

The first is the payment record. Regulation 38A of the Procurement Regulations 2024 came into force on 1 April 2026, and it says exactly what a contracting authority must publish about a payment: the paying authority’s name, postal address, email address and unique identifier; the same set again for the authority that published the contract details notice where that is a different body; unique identifiers for the procurement and the contract; the supplier’s name and unique identifier; the value net of VAT; and the date the payment was made. None of that is left to the authority to decide. Regulation 39, which governs the verdict on the supplier, has not been amended since February 2025.

The second is steel. Procurement Policy Note 022, updated in July 2026, tells central government buyers that from 1 October they should record in the contract details notice whether UK-produced steel will be used, on projects worth ten million pounds or more or expected to need more than five hundred tonnes of it. The guidance to that note does not leave the form of the record to anybody. Paragraph 13 supplies four strings and the box to type them in: [STEEL-ORIGIN: YES], [STEEL-ORIGIN: NO] followed by a rationale, [STEEL-ORIGIN: MIXED] followed by a rationale for the non-UK elements, and [STEEL-ORIGIN: NOT KNOWN] followed by a rationale. Two worked model texts are supplied so nobody has to invent one. Contractors owe their origin data to the buying organization within ten weeks of the financial year ending, on a standard template, and the organization consolidates it into an annual steel data return to the Department for Business and Trade covering quantity, product type and origin. Origin comes off the EN10204 Type 3.1 inspection certificate, and paragraph 73 asks the contractor to say whether that certificate records where the steel was melted and poured.

Paragraph 66 of that same guidance is what it has to say about whether the contract worked: “KPIs should drive both a focus on outcomes and continuous improvement (aligning with the Social Value Model where appropriate) and the intended benefits to be realised during contract delivery.” There is no field, no set of permitted values, no template, and nobody it goes to. That is the same document, written by the same department in the same year. It can trace a girder back to the furnace where it was poured, and on whether the contract delivered what it was bought for it asks for a focus on outcomes.

The First Notice Shows the Shape

Ward Hadaway reported on 27 February 2026 that the first contract performance notice under the Act had been published, covering a uniform contract for the University of Sheffield that ran a little over four months at £432,000 including VAT. It was rated good, the highest category available. They also flagged the detail that matters here, which is that the contract sat far below the five million pound threshold, so no obligation to set indicators applied to it at all. A public rating was published anyway.

Their interest was in whether that becomes a pattern. Mine is in what a reader learns from the notice. A supplier now carries a published good rating from a public body, and nothing on the outside can establish what it was good at or how that was evidenced. A rating means as much as the standard behind it, and the standard here was whatever that authority decided to write down.

Why This Matters Well Outside UK Public Procurement

I wrote in Buyers Judge Vendor Claims by Size and Ignore the Track Record that nothing rates the credibility of a vendor’s claim before the money moves. Britain has now built something close to a public rating of what happened afterwards, which is genuine progress and worth watching. It also shows where the difficulty sits, because constructing the publication machinery turned out to be the achievable half, and deciding what the publication has to be about is the half that determines whether any of it informs a reader.

That is the pattern from Colorado and the EU Kept AI Disclosure and Dropped AI Assessment running in the opposite direction. Colorado repealed its deployer assessment duty and the European Union deferred its high-risk obligations, while Britain added a publication duty that does not reach value. Adding and removing produce the same outcome for a buyer when neither says what has to be verified.

What a Buyer Does With This

If your contract falls inside the duty you already have the obligation and the platform, and what you decide the indicators measure is the whole of it. If it sits in one of the section 52(6) categories the duty never reaches you, which makes the record yours to build or not build, and building it is still the right call. Set at least one that names the outcome the contract exists to produce, define before signature what evidence would show whether it arrived and who supplies that evidence, and keep that definition in your own files rather than in the supplier’s, so the measure cannot move quietly after the first difficult quarter.

When I wrote that, it was a caution. Regulation 39 turns it into a documented route, and the only trace a reader gets is whatever somebody types into that free text box.

Everyone else should read this as a preview. The obligation to publish a verdict is spreading, and it will reach commercial buying long before anyone legislates what the verdict has to be about. The Act publishes whatever you choose to measure and has no opinion on whether you chose the right thing.

The European Commission’s September 2026 procurement proposal runs the other way: it mandates a quality weighting and asks for a record of each contract’s result, but builds no duty to check that result against the quality the award paid for.

The same government’s national value standard for medical technology, published in June 2026, stops at the same place from the other side: mandatory weightings at award, and no duty to record whether the value arrived.

What happens to the verdict after it is published is its own question, and I took it up in UK Buyers Now Rate Suppliers Publicly, With No Way to Compare: the notice is written for other contracting authorities, and the register that would carry it to them is still to be built.

See also Procurement Prices Every Concession and Never Checks the Result, on the concession-pricing apparatus and the verification line it leaves out.

Keep reading: ACT II · Chapter 11: Outcome Credibility Markets