A technically stronger candidate can still be the weaker educational purchase. The following is a synthetic calculation, not an assessment of real suppliers. Assume both candidates have cleared all applicable gates and have complete evidence after a pilot.
| Criterion | Weight | Candidate A: impressive builder, weak teaching | Candidate B: evidence-led mentor |
|---|---|---|---|
| Learner understanding and ownership | 22 | 1 | 4 |
| Teaching and feedback | 18 | 1 | 4 |
| Relevant technical competence | 15 | 4 | 3 |
| Readiness, scope and progression | 10 | 2 | 4 |
| Rapport, agency and accessibility | 10 | 3 | 3 |
| Evaluation and technical judgment | 8 | 1 | 3 |
| Parent reporting | 7 | 1 | 3 |
| Continuity and coordination | 5 | 3 | 3 |
| Commercial clarity | 5 | 2 | 3 |
| Weighted total | 100 | 47.5 | 87.5 |
Candidate B’s ratings of 4 assume repeated supporting observations. They cannot be awarded for a persuasive first consultation.
Candidate A is not recommended for the longer engagement despite excellent technical competence. Candidate B is a stronger purchase under this methodology, provided the child wants to continue and the next block remains within the demonstrated scope.
A third supplier with an excellent proposal but no named-mentor observation should be shown separately as incomplete. Do not fill its unknowns with average scores to make the shortlist look finished.
Sources and context
This page sets out the desk’s purchasing framework. What are the limits of this research? → Sources →