Scorecard · No. 15 / 17
What does a completed score look like?
Last updated
The total is a decision aid; the gates and critical floors determine whether it is usable. The following is a fictional calibration example, not a supplier assessment.
| Criterion | Weight | Hypothetical rating | Weighted points |
|---|---|---|---|
| Commercial decision and fit | 10 | 4 | 8.0 |
| Buying prompts | 12 | 4 | 9.6 |
| Surface fidelity | 8 | 3 | 4.8 |
| Sampling and comparability | 14 | 4 | 11.2 |
| Metric definitions | 10 | 4 | 8.0 |
| Evidence and QA | 10 | 4 | 8.0 |
| Diagnosis | 10 | 3 | 6.0 |
| Delivery | 8 | 4 | 6.4 |
| Commercial measurement | 6 | 3 | 3.6 |
| Accuracy | 4 | 4 | 3.2 |
| Access and security | 4 | 4 | 3.2 |
| Cost and contract | 4 | 4 | 3.2 |
| Total | 100 | 75.2 |
Assuming all gates pass, this arrangement qualifies for a conditional pilot. Its conditions should address the weaker surface specification, diagnostic evidence and commercial measurement plan—not merely request a higher headline score.
For contrast, an otherwise perfect proposal scoring only 2/5 on sampling and evidence access would still total 85.6/100. It nevertheless fails the critical floors. This is why a weighted average alone is unsafe as an approval mechanism.
How should the buyer interpret the total?
| Total | Proposed decision | Conditions |
|---|---|---|
| 85–100 | Strong shortlist candidate for a bounded pilot | All gates pass; critical floors met; no unresolved dependency that defeats the purchase |
| 70–84.9 | Conditional pilot | Written conditions, owners and evidence deadlines attached |
| 55–69.9 | Revise or narrow the proposal | Resolve the main evidence or execution gaps before programme approval |
| Below 55 | Do not approve the proposed programme | Consider a smaller diagnostic, an existing supplier or no new purchase |
| Any total with a failed gate or critical floor | Do not approve affected work | Resolve the issue and rescore; do not trade it against strengths elsewhere |
A new plan with documentation but no demonstration will naturally score lower than a tested arrangement. Authorising a small procurement validation exercise is different from approving the full programme.