Your tender is not read. It is scored.
That difference changes how you should write. A reader forms an impression. An evaluator works through a marking scheme, question by question, and must justify every mark in writing. I have done this work as a buyer. Here is how tenders are evaluated, and what it means for the way you write.
Each answer is scored against a descriptor
Most quality questions are marked on a published scale, often 0 to 5. Each score has a descriptor. A typical top descriptor reads something like: a full response, with detailed evidence, that gives the authority high confidence in delivery. Lower descriptors describe partial answers, generic answers, or answers with gaps.
The evaluator's task is to match your answer to the descriptor that fits it best. They are not deciding whether they like you. They are deciding which description your answer meets.
What to do: Find the scoring descriptors in the tender pack before you write. Read the top one. Write an answer that meets every part of it, and make each part easy to see.
Every mark has to be justified in writing
Evaluators record a reason for each score. That record may be shared with bidders and may be scrutinised if a decision is challenged. So evaluators give marks only for what they can point to on the page.
This is where long, flowing answers lose. If the evidence for a point is buried in the fourth paragraph, the evaluator may not find it. If it is not there at all, they cannot credit it, however good your service really is.
What to do: Mirror the structure of the question. If the question has four parts, use four subheadings in the same order. Put your strongest evidence in the first two lines of each part.
Scores are agreed in a moderation meeting
Evaluators usually score on their own first. The panel then meets to agree a single score for each answer. Differences are discussed and settled.
In that meeting, ambiguity works against you. If one evaluator read your answer generously and another did not, the panel will settle on what the page clearly supports. Nobody in the room can speak up for what you meant.
What to do: Remove any sentence that could be read two ways. State commitments plainly: who does what, by when, measured how.
Weighting decides where your effort goes
Not every question carries the same weight. A question worth 20% of the quality score deserves more of your time than one worth 5%. Many bidders spend equal effort on every answer, or the most effort on the questions they find easiest.
Under the Procurement Act 2023, contracts are awarded to the most advantageous tender, judged against the award criteria and weightings set out at the start. The weighting tells you exactly what the buyer values.
What to do: Rank the questions by weighting before you start. Draft the heaviest ones first, while you are fresh.
An illustrative example
The question reads: describe how you will manage complaints, including timescales, escalation and learning.
An answer written for a reader tells a story about the firm's customer focus, then mentions complaints in passing.
An answer written for an evaluator uses three subheadings: Timescales, Escalation, Learning. Under each, it gives the commitment, the named role responsible, and one piece of evidence from a current contract.
The second answer can be scored in two minutes. The evaluator can tick every part of the descriptor and write the justification without searching.
This example is illustrative. It is not taken from a real bid.
Write for the person holding the marking scheme
The evaluator wants to give you marks. Their job is to find the evidence that lets them. Your job is to make that easy.
Write for the scorer, not the reader.
Want to see your last bid through an evaluator's eyes? Grow Smat offers a free bid teardown for firms that have recently lost a tender. Request your free teardown
Sat on an evaluation panel and seen it differently? Or have a question this post did not answer? Join the discussion on LinkedIn, or email [email protected].