The fastest way to understand how your bid will be marked is to see the instrument that marks it. A scoring matrix is no more than a structured way of turning a written answer into a number — but most suppliers never look at one closely, and so they write blind. This article walks through the components and gives worked examples you can hold your own responses up against.
For the behaviour behind the matrix — how evaluators actually use it — start with how procurement panels score quality.
The three parts of a scoring matrix
1. Criteria and weightings. The matrix lists each award criterion and the share of total marks it carries. This is where the contract is really won. A simplified example:
| Award criterion | Weighting |
|---|---|
| Quality of methodology / delivery approach | 35% |
| Relevant experience and case studies | 20% |
| Risk management | 15% |
| Social value | 10% |
| Price | 20% |
Note what this tells you instantly: methodology and experience together carry 55% of the marks. An hour spent strengthening those is worth far more than the same hour on social value or price.
2. The band scale. Each qualitative criterion is scored on a banded scale. A common 0–5 scale, with descriptors:
| Score | Band | What it means |
|---|---|---|
| 5 | Excellent | Fully meets the requirement; comprehensive, with strong, specific supporting evidence |
| 4 | Good | Meets the requirement well; mostly evidenced, minor gaps |
| 3 | Satisfactory | Meets the requirement but with limited or generic evidence |
| 2 | Minor concerns | Partially meets the requirement; notable gaps or unsupported claims |
| 1 | Serious concerns | Largely fails to meet the requirement |
| 0 | Unacceptable | Does not address the requirement / non-compliant |
3. The calculation. Your band score on each criterion is converted to a percentage of that criterion’s available marks and multiplied by its weighting, then summed. A 4/5 on a 35%-weighted criterion contributes 0.8 × 35 = 28 of a possible 35 marks.
Where the band descriptors do their damage
Look hard at the gap between a 3 and a 4 above. It is not “better writing.” It is evidence. “Satisfactory but with limited or generic evidence” versus “meets the requirement well, mostly evidenced.” The thing standing between those two bands — and between a competitive bid and an also-ran — is whether the evaluator has something specific to point to.
This is the generic ceiling in matrix form: a complete, compliant, well-written answer with no proof points lands at a 3. It cannot reach a 4 on prose alone, no matter how good the prose is. (We unpack the most common reasons answers stall in why bids lose marks.)
A worked example: the same answer, scored twice
Criterion (35% weighted): Describe your approach to minimising service disruption during transition.
Version A — scores a 3 (satisfactory): “We have a proven, robust transition methodology refined over many years. Our experienced team follows industry best practice to ensure a smooth handover with minimal disruption to your service. Quality and continuity are our highest priorities throughout.”
Reads fine. Says nothing scoreable. No named comparison, no figure, no evidence the evaluator can defend a high mark with. Mid-band.
Version B — scores a 5 (excellent): “We will run transition in three gated phases over eight weeks, with the legacy service kept live in parallel until each phase passes acceptance testing. On the [Local Authority X] transition (a contract of comparable scale, 2024), this approach delivered zero unplanned downtime and a 99.98% service availability rate through the cut-over period — verifiable against the reference at Annex C. A named transition lead is accountable for each gate, with a defined rollback plan if any phase fails acceptance.”
Same capability. The difference is entirely evidence and specificity: a defined method, a named comparable contract, a measurable outcome, a verifiable reference, clear accountability, and a stated risk response. The evaluator can now write “5 — comprehensive, strongly evidenced” and defend it in moderation.
How to use a matrix on your own bid
- Rebuild the matrix from the tender — criteria, weightings, band descriptors, thresholds.
- Score each of your answers against the band descriptors, honestly, in the evaluator’s words.
- For any answer below a 4 on a high-weighted criterion, ask: what evidence would move it up a band?
- Identify any minimum thresholds and confirm every answer clears them.
- Reallocate your remaining effort toward the highest-weighted, lowest-scoring answers.
Done properly, this is a structured discipline rather than a guess — the basis of a proper bid review methodology.
Score your bid against the real matrix — automatically. BidReview marks your live response criterion by criterion, in band language, and shows where evidence would move you up. Run the free scorecard →
