How to write a tender response: a UK construction guide

The Invitation to Tender (ITT) is the marking scheme. Treat it as such and the writing becomes far easier — you are answering a known question to a known scale, not composing free-form prose. Before drafting anything, extract and list:

• Each scored criterion and its exact published wording.
• The weighting for each criterion (quality, price, social value, etc.).
• The scoring scale — 0–5 with band descriptors, 0–100 percentage, or a pass/fail threshold.
• Any minimum or pass mark that must be met to avoid elimination.
• The word or page limit for each answer.
• Any mandatory returnables, certificates, or forms.

Two of these cost more marks than any other. First, the scoring scale: a 0–5 scale with named bands rewards different behaviour to a 0–100 percentage scale. Evaluators mark to the band descriptors, so your answer must contain the sentence that justifies the band you want. Second, the word limit: text beyond a stated limit is routinely not read, so the strongest evidence must sit inside the cut, never after it.

Under the Procurement Act 2023, authorities must publish their award criteria and weightings in the tender notice, so this information is available — use it. Writing to the old "MEAT" language in 2026 signals a recycled bid library.

Strong quality answers share a consistent four-part structure. Each part maps to something an evaluator can tick against the band descriptors:

1. What you will do — the approach or method, stated directly in response to the criterion wording.
2. How it works on this project — the method applied to the specific site, programme, risks and client, not a generic statement.
3. Proof it has worked before — named projects, quantified outcomes, references, and data that confirm the approach is real and delivered.
4. How it is governed and measured — who owns it, how it is monitored, and what happens if it goes wrong.

This structure works because it mirrors how evaluators score: they look for method, project-specific application, evidence, and assurance. If any of the four is missing, the answer drops a band. If all four are present and specific, you give the evaluator no reason to mark below the top band.

The most common reason a well-written answer scores mid-band rather than top is missing evidence. Describing an approach without proving it has been delivered tells the evaluator the method is plausible but unconfirmed. Many criteria explicitly ask for both method and evidence, and they are scored as two separate things.

Strong evidence is concrete and quantified:

• A named project with a comparable scope and value.
• A specific outcome — "reduced defects by 40%", "delivered 95% on programme", not "excellent results".
• A reference or testimonial the evaluator could verify.
• A named team member responsible for delivery.

Source your evidence before you start writing. Anything only your business can supply — actual project data, references, named staff, certifications — must be gathered at the planning stage. Trying to find a reference number two days before submission is how otherwise strong answers collapse into unsupported assertion.

When an answer has a 500-word limit and you have 700 words of good material, the discipline is not "trim 200 words" — it is "decide which evidence earns the top band and put that first." Evaluators routinely do not read past the stated limit, so anything beyond the cut is wasted.

A practical approach: write the answer, then score it against the band descriptors yourself. Identify the sentences that justify the band you want. Those go in the first paragraph. Supporting detail follows. Anything that does not help the evaluator award a higher band is cut, no matter how proud you are of it.

The same applies to page limits. If the ITT says "maximum 4 pages", do not treat it as a target. A focused 3-page answer that hits every band descriptor beats a padded 4-page answer that buries the evidence on page 4.

Social value is now its own discipline, not an appendix to sustainability. Under the Procurement Act 2023's transparency regime, performance against social value commitments is published — so vague pledges are increasingly hard to defend.

Social value scores when it is:

• Specific — a number, not a sentiment ("12 apprenticeships on this contract", not "we support apprenticeships").
• Local — tied to the project's community, not a national policy statement.
• Measurable — with a reporting mechanism and a baseline.
• Owned — a named person accountable for delivery.

The TOMs (National TOMs Framework) is a common measurement system, but check whether the ITT specifies its own social value framework — many authorities do. Write to that framework's measures, not a generic TOMs menu.

Under the Procurement Act 2023, award is on the Most Advantageous Tender (MAT), not the cheapest. Authorities publish the quality/price weighting, so you know exactly how much each element matters. A 60/40 quality/price split rewards a strong quality submission; a 30/70 split means price competitiveness dominates.

The practical implication: your pricing strategy must be consistent with your quality narrative. If you claim a premium approach with exceptional quality management, your price should reflect that — a low price with a high-quality narrative creates credibility problems. Conversely, if you are competing on price, make sure your quality answers meet the minimum threshold rather than over-investing in top-band prose that the weighting does not reward.

Match your price submission to the pricing schedule exactly. A mismatch between the narrative and the numbers is a red flag for evaluators and can trigger clarification questions that delay or derail the evaluation.

The review stage is where most bids gain or lose their final band. Review against the ITT's own methodology — the criteria, the scale, the band descriptors — not against whether the prose "reads well". A reviewer should be able to:

• Award a band to each answer and quote the sentence that justifies it.
• Confirm every scored criterion has an answer, in the ITT's order and numbering.
• Confirm every answer is inside its word or page limit, with evidence before the cut.
• Confirm every claim has a source — a project, a number, a named commitment.
• Check for placeholders, residue from another bid, and wrong client names.
• Verify weightings total 100% and the price submission matches the schedule.
• Confirm all returnables, certificates and forms are attached in the required format.

If the reviewer cannot find the sentence that justifies a band, the mark is not there. That sentence either needs to be written or the answer needs to be restructured so the evidence is visible.

The structural problems that lose marks are remarkably consistent across UK construction tenders:

• Answering the wrong question. Writing a strong answer to a criterion that was not asked, while the actual criterion goes under-addressed. Always answer the published wording.
• Method without evidence. A well-described approach with no proof of delivery. Evaluators cannot confirm the method is real, so it scores mid-band.
• Generic social value. The same corporate policy paragraph in every bid. It answers nothing the evaluator can tick and scores low almost everywhere.
• Evidence beyond the word limit. Strong material sitting after the cut, where it is never read. Put your best evidence first.
• Bid library residue. A previous client's name, an old framework reference, or a project that does not match the current scope. These signal a recycled bid and undermine credibility.
• Inconsistent pricing. A quality narrative that promises a premium approach with a price that suggests corner-cutting. Align the two.

Every one of these is preventable at the planning and review stages — which is why disciplined bid management raises win rate more than better writing does.

How do I start writing a tender response?

Before writing a single answer, extract every scored criterion from the ITT — its exact wording, weighting, the scoring scale and band descriptors, any minimum or pass threshold, and the word or page limit for each response. Your answers are then written against those criteria, in the order they appear, with evidence positioned before any word-limit cut. Starting to write without this extraction is the single most common cause of lost marks.

How long should a tender response be?

As long as the ITT allows and no longer. Most UK construction tenders set a word or page limit per criterion, and text beyond that limit is routinely not read by evaluators. The discipline is fitting your strongest evidence inside the cut, not filling the limit for its own sake. If the ITT does not state a limit, write enough to answer the criterion fully without padding.

What makes a tender response score well?

High-scoring answers do three things: they answer the published criterion wording directly, they back every claim with evidence — named projects, numbers, references — and they show how the approach works on this specific project rather than recycling a generic method statement. Evaluators mark to band descriptors, so the sentence that justifies a band must actually exist in the answer.

How do I write social value in a construction tender?

Social value scores when commitments are specific, local, measurable and owned. That means apprenticeship numbers tied to the contract value, named local supply-chain spend targets, a named person accountable for delivery, and a reporting mechanism. Generic corporate policy statements — the same paragraph in every bid — score low because they answer nothing the evaluator can tick.

What is the difference between method and evidence in a tender?

Method is what you will do and how it works; evidence is proof that it has worked before, with named projects and quantified outcomes. Many criteria ask for both, and they are scored as two separate things. An answer that describes a robust approach but provides no proof of delivery typically lands mid-band at best, because the evaluator cannot confirm the method is real.

Should I use the same answer for every framework?

No. Each UK framework — Pagabo, Procure Partnerships, SCF, YPO, NHS SBS, Crown Commercial Service — sets its own call-off mechanics, scoring scales and evidence expectations. An answer written for one framework's rubric and dropped into another scores as generic because it answers a question that was not asked. Rewrite against the call-off documentation each time.