When tender answers contradict each other — and how to catch it before submission

A contradiction is any place where two statements in the same bid cannot both be true at the same time, or where a statement in the bid conflicts with a requirement the authority has published. The common forms in UK construction tenders are:

• Programme vs method. The method statement describes a sequenced build over 42 weeks; the programme answer states 36 weeks to a key date. Both cannot be the plan. An evaluator reading both will not award the higher band to either.
• Numbers that disagree. The social value answer commits to 12 apprenticeships; the training and employment answer commits to 8. The mobilisation answer carries a different number again. The evaluator can only believe one figure.
• Named people vs roles. One answer names a project director with a CV in the returnables; another answer implies a different person holding the same role. The evaluator does not know who is accountable.
• Requirement vs answer. The ITT asks for a CDM principal designer appointment evidenced by a signed letter; the answer describes the role but never mentions the letter. The answer is not non-compliant, but it discharges less of the ask than the band descriptor requires.
• Answer vs clarification. The authority publishes a clarification moving the site start date; a programme answer still uses the original date. This is the most dangerous form, because the bidder has the corrected information and ignored it.

None of these is a writing-quality problem. They are consistency problems, and they are invisible to the person who drafted either answer in isolation.

UK public-sector evaluation is criterion-based and evidence-led. Each scored criterion carries published band descriptors, and evaluators are instructed to award a band only where the evidence in the answer meets the descriptor's standard. The descriptor is not satisfied by an answer that is internally inconsistent.

When two answers contradict, one of three things happens, and none is good for the bidder:

• The evaluator awards the lower band to both, on the principle that a bidder who cannot keep its own answers consistent has not demonstrated control of the subject matter.
• The evaluator raises the contradiction as a reservation or clarification point in moderation, which can pull the whole quality score down under a moderation adjustment.
• On a compliance criterion — a pass/fail or threshold requirement — the contradiction can trigger a fail, because the evaluator cannot confirm the bidder meets the requirement as clarified.

The loss is rarely a single band on a single criterion. A contradiction that touches programme, method and mobilisation can pull three criteria down at once, and the weighted effect on the overall quality score is larger than the mark on any one of them.

Contradictions are a process failure, not a competence failure. The conditions that create them are normal in bid management:

• Parallel drafting. Several authors write different criteria at the same time, each working from their own reading of the ITT. Nobody owns the whole answer set until review, by which time the contradictions are baked in.
• Reused content. Text lifted from a previous bid carries the previous project's programme, numbers and named people. If it is not checked against the current ITT, it contradicts the answers written fresh for this tender.
• Clarifications arriving late. The clarification window opens before drafting finishes. A clarification that changes a date, a scope boundary or a methodology arrives after some answers are already written, and those answers are not always revisited.
• Siloed teams. The commercial team owns price, the delivery team owns method and programme, the bid manager owns the narrative. Each is internally consistent; the contradictions only appear where their answers meet.
• No single source of truth. When the project facts live in heads, inboxes and spreadsheets rather than in one register, two authors can each write a different, honestly-held figure.

A late-stage review can catch the obvious cases, but the review is usually done under time pressure, by the same people who wrote the answers, looking for typos. Contradictions that sit across criteria are exactly what a tired human review misses.

Under the Procurement Act 2023, clarification answers published by the contracting authority can amend the ITT. A clarification that moves a date, adds a returnable, or changes a methodology has the same standing as the original document. A response that satisfies the original wording but ignores the clarification is treated as non-compliant with the clarified requirement.

This is the highest-risk contradiction, because the bidder holds the corrected information and still submitted the stale answer. Evaluators notice it quickly — they wrote the clarification — and it reads as a failure to read the documents properly, which is exactly the impression a quality criterion is designed to test.

The defence is to treat every published clarification as a first-class source document, not a footnote, and to check every answer against the clarification log before submission. That check is mechanical enough to be done consistently, but it is almost never done consistently by hand.

TendersIQ UK is built around a project fact register and a cross-check that runs before submission. The flow is:

1. Read the documents. TendersIQ reads every uploaded tender document — the ITT, the clarification log, the pricing schedule, the returnables list — and builds a register of the facts the bid has to be consistent with: dates, numbers, named roles, methodologies, scope boundaries.
2. Rebuild the scoring framework. The published criteria, weightings and band descriptors are reconstructed criterion by criterion, so every answer is marked against the authority's own methodology, not a generic quality narrative.
3. Score each answer. The AI classifies each answer against its criterion; a deterministic scorer assigns the band. This keeps the scoring honest — no model-emitted number reaches the score.
4. Cross-check the answers. The cross-check compares every answer against the fact register and against every other answer. Where a statement in one answer conflicts with a fact, a clarification, or a statement elsewhere in the bid, it flags the conflict with the quotes on both sides.
5. Surface it before submission. Flagged conflicts appear in the pre-submission panel, so the team can resolve them — change one answer, reconcile the numbers, or confirm the clarification was applied — before the bid is submitted, not after the evaluator reads it.

The cross-check is the piece a manual review cannot do reliably: it reads the whole answer set against the whole document set, and it does it every time. The contradictions it catches are the ones that would otherwise cost bands across several criteria at once.

Read more about the engine and the honesty guardrails on the accuracy page, and see it run end to end on a demonstration tender in the guided walkthrough.

If you are reviewing a bid by hand, the contradictions most worth checking are the ones that cross team boundaries. Work through these before the bid goes out:

• Every date in every answer matches the programme and the clarification log — site start, key dates, handover, defects period.
• Every number that appears in more than one answer is the same number — apprenticeships, local spend, working hours, supervision ratios, tonnage, floor area.
• Every named person appears once, in the right role, and matches the CV in the returnables.
• The method statement describes the same build sequence and duration as the programme answer.
• Every requirement amended by a clarification is reflected in the answer that addresses it.
• The pricing schedule and the price answer carry the same totals and the same breakdown.
• Social value commitments match the answers that score them, and are specific, local and measurable — not carried over from a different project.

A checklist catches the contradictions you remember to look for. A cross-check that reads the whole bid catches the ones you don't.

What are tender queries in UK construction procurement?

Tender queries — sometimes called clarification questions — are the questions bidders submit to the contracting authority during the clarification period set out in the ITT. The authority publishes answers to all bidders, usually anonymously, as a clarification log. Under the Procurement Act 2023 these answers can themselves amend the ITT, so a bidder's response must be consistent with the published clarifications, not just the original documents.

How do contradictory answers lose marks?

Evaluators score each criterion against its published band descriptors. When one answer commits to a programme, a method or a number that another answer contradicts, the evaluator cannot award the higher band without ignoring the inconsistency — and most scoring methodologies instruct them not to. The contradiction reads as a lack of internal control, and the weaker answer drags the stronger one down with it.

Why do bid teams produce contradictions?

Most contradictions are not deliberate. They happen when several people write in parallel, when text is reused from a previous bid without checking it against the current ITT, when a clarification answer changes a requirement after an answer was drafted, or when the programme team and the method team work from different assumptions. The contradiction is rarely visible to the person who wrote either answer.

How does TendersIQ UK catch contradictions before submission?

TendersIQ reads every tender document and published clarification, rebuilds the scoring framework criterion by criterion, and cross-checks the answers against each other and against the source documents. Where a statement in one answer conflicts with a statement elsewhere, or with a clarification that amended the requirement, it flags the conflict before submission — with the quotes on both sides so the team can resolve it.

Is a clarification answer binding on bidders?

Yes. When a clarification answer amends a requirement, a timescale or a methodology, it has the same standing as the ITT. A response that ignores a published clarification is treated as non-compliant with the clarified requirement, even if it satisfied the original wording. TendersIQ treats clarification answers as part of the source documents and checks every answer against them.