PPP Consultants
Bids are scored, not read — writing infrastructure tenders to the evaluation matrix
An evaluation panel does not form a view of the best bid. It applies published criteria, line by line, and records a score. The bid written to that method wins marks the narrative bid never competes for.
PPP Consultants · September 2026
When a consortium delivers its bid for a Public Private Partnership (PPP) concession or a large engineering, procurement and construction (EPC) contract, the people who wrote it imagine the document being read. The evaluation panel opens the bid, follows the argument and forms a view. That is not what happens.
The panel scores the bid against a published evaluation matrix — a table of criteria, each with a weighting, each with a scale of marks — and the award goes to the highest total.
A bid structured to that matrix has a clear advantage over a bid structured to the bidder’s preferred story, even where both are technically equal. This article sets out the discipline that produces the first kind of bid, and the evidence that it is worth the effort.
How a bid is scored.
Public procurement law in most jurisdictions now requires the scoring method to be published before bids are invited, and requires the panel to apply it as published. The United Kingdom’s Procurement Act 2023, in force since 24 February 2025, is a clear example.
Section 23 requires that award criteria be “sufficiently clear, measurable and specific”, that they relate to the subject-matter of the contract, and that the authority describe how tenders are to be assessed against them — including whether failure on any criterion disqualifies a tender — and state the relative importance of each criterion, by percentage weighting or by ranking. Section 19 then obliges the authority to award to the tender that “best satisfies the award criteria when assessed by reference to the assessment methodology”.
The World Bank’s procurement regulations for the projects it finances, and the procurement rules of most Gulf and African PPP regimes, work the same way: rated criteria, published weightings, recorded scores.
The consequence for the bidder is precise. The panel is not permitted to reward what the criteria do not ask for, and it is not permitted to overlook what they do. Each evaluator holds one or more criteria and a scale that says what earns each mark. The evaluator reads the bid looking for the evidence that fits the scale, records a mark and a reason, and moves on.
A page that does not map to a criterion contributes nothing. A page that answers a criterion but places the evidence where the evaluator is not looking is marked down or missed. The bid’s task is therefore not to persuade but to make every mark easy to award.
Why it matters more than it used to.
Three things have raised the stakes. First, competition for the bids that are worth winning has thinned. The United Kingdom’s National Audit Office found that the share of PPP projects attracting three or more developed bids fell from 85 per cent before 2004 to 67 per cent for projects closing between 2004 and 2006, as bidders learned how long and expensive the process had become.
Fewer bidders mean larger, costlier bids from those who remain, and a higher price for losing. Second, panels are more international. A Gulf ministry, an African parastatal or a European development bank may seat evaluators for whom English is a second or third language, and who will score consistently only if the evidence is written so that it cannot be misread.
Third, the published method now carries legal weight: an authority that departs from its own scoring can be challenged by a losing bidder, so panels keep to the matrix with some care. The bidder who writes to the matrix is writing to the one thing the panel is bound to follow.
The matrix as the brief.
The published evaluation criteria attached to a Request for Proposals are the bid’s true brief. Not the project description, not the technical scope, not the language of the covering letter — the criteria. Each criterion has a weighting. Each section of the response will be assessed against one or more criteria. Sections that map to no criterion earn no marks.
The discipline that follows is structural, and it begins before a page is drafted.
The bid team takes the evaluation matrix apart and builds a response architecture in which every criterion is answered by a defined section of the bid, in proportion to its weighting: a criterion worth 20 per cent of the technical score receives the attention, the evidence and the pages that 20 per cent deserves, and a criterion worth 2 per cent receives what 2 per cent deserves and no more.
The bid is then written to fill that architecture. Where the matrix is ambiguous — and matrices often are — the team reads it as the panel will read it, not as the bidder would like to, and asks a clarification question if the ambiguity is material.
Writing to the criteria, not to the story.
Bidders who write to the criteria and bidders who write to a story produce documents that look similar and score differently. The story-led bid develops an argument across its sections; the criteria-led bid gives each evaluator, at the point in the document where the evaluator expects it, the evidence that earns the top mark for the criterion in front of them. When the two meet a panel scoring against a matrix, the criteria-led bid wins marks the story-led bid did not know it was competing for.
Three practices follow. Every section opens by naming the criterion it answers, in the authority’s own words, so that the evaluator knows at once that the evidence is here. Evidence is stated as evidence — the project, the date, the value, the outcome, the reference — and not as assertion. Where the same evidence supports two criteria, it appears in both places; the panel does not cross-refer on the bidder’s behalf, and a mark not scored is a mark lost.
The English the bid is written in.
International procurements routinely receive bids from consortia working in English as a second language and are routinely scored by panels doing the same. The technical content may be excellent. Where the panel is scoring in English against a matrix, the bid written in clear, evaluable English carries an advantage that compounds across every criterion, because the evaluator recognises the evidence without re-reading.
This is not a question of literary polish. It is a question of short sentences, one point per paragraph, the criterion’s own vocabulary, numbers rather than adjectives, and the removal of every word that does not help the evaluator award a mark. A bid written this way reads, to the consortium that produced it, as somewhat over-engineered: more signposting than feels natural, more repetition of evidence, less ambition in the prose. Those are the properties that score.
Why it pays twice.
Where the project will be financed by lenders, the way risk is presented in the bid has a second effect after the award. Lenders price the risk they perceive. A bid documented with transparent, evidenced risk allocation is not only a stronger bid on the matrix; it is the first document a lender’s analyst reads, and it sets up the financing terms that follow.
The discipline that wins the marks also lowers the cost of capital.
A case from the firm’s work.
A European consortium — NYOP — sought to enter the United Kingdom PPP market, then the largest and most competitive in the world, as a newcomer to that market.
The firm’s Chief Executive led the bid as bid director for the first European-led joint venture into that market: assembling the consortium, mapping its capability to the evaluation criteria before partners were committed, directing the bid documentation and financial model, and leading the negotiation.
The bid was for a six-academy schools programme in Aberdeenshire, Scotland, of more than US$400 million in capital value. The consortium was selected as preferred bidder, reached Financial Close and all six academies are now built and operational. The joint venture went on to win further United Kingdom projects together. The engagement is on the public record.
The lesson is the one this article makes: a bidder with no local history won a national procurement because its bid was built to the matrix that scored it.
What to do differently.
A bidder preparing for a scored procurement should treat the evaluation matrix as the first document to be read and the last to be put down.
Build the response architecture from the matrix before drafting. Assign each criterion an owner, a page budget in proportion to its weighting and a list of the evidence that earns the top mark. Write in the panel’s language and to its scale. Have the draft scored, criterion by criterion, by someone who did not write it, and rewrite wherever the evidence could not be found at once.
That is the whole method. It is unglamorous, and it is how bids are won.
Sources: Procurement Act 2023 (UK), sections 19 and 23, legislation.gov.uk. National Audit Office, Improving the PFI tendering process, HC 149 Session 2006–07, March 2007. World Bank, Procurement Regulations for IPF Borrowers (rated criteria).