Scoring Criteria
All elements present. Reader who reads only the summary understands program, evaluation, headline conclusions, and top recommendations well enough to make decisions.
Reader can make most decisions from the summary alone. One or two specific points would require the full report.
Reader gets the gist but needs the full report to make confident decisions. Context partial.
Summary depends on the full report for basic comprehension. Jargon or assumed prior knowledge.
Summary is unintelligible as stand-alone. Reader cannot identify what was evaluated.
Top three to five findings surfaced within the first read, visually distinguished, ordered by importance. Secondary findings subordinated.
Top findings identifiable on first read. Ordering or visual treatment partial.
Findings flattened. All findings receive similar weight.
Headline findings buried among secondary observations. Reader could miss them.
No salience structure. Findings absent, scattered, or too generic.
All five elements present in proportionate space: purpose, methods in brief, key findings, conclusions, recommendations.
Four of five elements present and proportionate. One element light but not absent.
Three of five elements present. One or two absent or too brief.
Two of five present. Findings or conclusions dominate; others missing.
One element present, or coverage so uneven the summary fails as an overview.
Accessible to a non-specialist reader. Acronyms defined or avoided. Technical terms translated. Short, active sentences.
Mostly accessible. One or two technical terms or acronyms unexplained but not obstructive.
Mixed register. Some sections clear, others jargon-heavy. Inconsistent acronym definitions.
Heavily technical. Multiple undefined acronyms or methodological jargon.
Reads as a methods chapter. Inaccessible outside the evaluation team.
Appropriate length (2 to 4 pages for major evaluation, 1 to 2 for smaller). Condensed, not excerpted.
Within one page of target. Minor padding or repetition.
Slightly off-target. Either compressed enough to lose elements or expanded with non-essential detail.
Clearly off-target (under 1 page for major, or 6 to 8 pages). Thin or padded.
So off-target the summary fails its purpose.
Score Interpretation
| Total (out of 25) | Band | Next Step |
|---|---|---|
| 22-25 | Strong | Summary is ready to serve as a decision-maker's brief. Approve. |
| 17-21 | Adequate | Address flagged dimensions before circulating. Most likely fixes: tighten finding salience or restore a missing coverage element. |
| 11-16 | Needs Revision | Return to the drafter with the scorecard as revision brief. Most common gap is stand-alone value or coverage completeness. |
| 5-10 | Substantial Revision | Summary does not function as an executive summary. Rebuild from the full report using the Generate or Revise prompt. |