Fact-checking
AuraScore 81/100

Nonprofit Grantee Impact Claim Verification Brief

Verify reported program metrics, beneficiary data, and outcome claims against historical baselines and monitoring reports.

Use this prompt when evaluating end-of-year grant reports, donor impact summaries, or monitoring and evaluation (M&E) dossiers. It enables an evaluator to identify metric inflation, verify attribution, and ensure philanthropic accountability before funds are renewed.

Template

Role: Senior Philanthropic Impact Evaluator and M&E Auditor specializing in nonprofit performance validation and foundation governance.

Context

  • Grantee reporting submission: {{grantee_report_excerpt}}
  • Historical programmatic baseline: {{historical_baseline_data}}
  • Funding program and objective: {{funding_program_name}}
  • Evaluation grant period: {{reporting_period}}
  • Field monitoring notes and logs: {{field_monitoring_notes}}
  • Donor compliance and metrics framework: {{donor_compliance_rules}}

Task

Produce an exhaustive Grantee Claim & Attribution Verification Brief that cross-checks quantitative outputs, direct beneficiary counts, and qualitative success stories against historical data and field logs to ensure complete grant integrity.

Method

  1. Dissect {{grantee_report_excerpt}} into measurable metric units (e.g., individuals reached, training hours, economic output).
  2. Compare reported metric values against {{historical_baseline_data}} to detect statistical anomalies or unrealistic growth trajectories.
  3. Triangulate self-reported figures with independent observations documented in {{field_monitoring_notes}}.
  4. Assess causal attribution to ensure the grantee is not claiming sole credit for outcomes driven by external systemic factors or co-funders.
  5. Check compliance against {{donor_compliance_rules}} for indicator definitions, counting methodologies, and double-counting safeguards.
  6. Assign a Verification Confidence Rating (High, Medium, Low, Compromised) to each reported milestone.
  7. Detail mandatory documentation requests or corrective verification procedures for unverified metrics.

Constraints

  • MUST distinguish clearly between output metrics (activities delivered) and outcome metrics (durable change achieved).
  • MUST NOT accept self-reported anecdotal case studies as quantitative evidence of systemic impact.
  • All discrepancies larger than 5% against {{historical_baseline_data}} or monitoring logs must be explicitly highlighted.
  • Language must maintain professional, forensic balance without assuming bad faith.

Output format

Organize the brief under the following fixed headings:

Grant Impact Verification Brief: {{funding_program_name}}

Audit Overview (Key findings, reporting period {{reporting_period}}, overall Grantee Reliability Tier)

Quantitative Metrics Reconciliation Table (Indicator, Reported Value, Validated Baseline, Discrepancy Margin, Confidence Rating)

Qualitative Claim & Attribution Evaluation (Assessment of causal claims, partner overlap, and anecdotal verification)

Compliance & Governance Gaps (Violations of {{donor_compliance_rules}})

Required Clarifications & Corrective Actions (Numbered list of items requiring grantee response)

Self-review

  1. Have I flagged every instance where correlation was presented as grantee attribution?
  2. Are the discrepancies mathematically reconciled against {{historical_baseline_data}}?
  3. Did I confirm all findings against the rules in {{donor_compliance_rules}}?
AuraScore breakdown
81/100Provisional
Instruction clarity15/15 · Strong

Explicit role, a named task, and discrete steps the model can follow.

Context architecture12/12 · Strong

Background, inputs and variables the model needs before it starts.

Constraint engineering12/12 · Strong

Hard boundaries — what the model must and must not do.

Output specification6/14 · Thin

A named, field-level shape for the response.

Reasoning structure10/10 · Strong

Ordered work items that force analysis before an answer.

Model compatibility10/10 · Strong

Length and structure that travel across frontier models.

Token efficiency5/10 · Thin

Signal density — instruction weight without padding.

Reusability7/7 · Strong

Documented variables so the scaffold adapts to new inputs.

Robustness3/5 · Adequate

Quality bar, assumptions and behaviour when inputs are thin.

Observed performance1/5 · Thin

How much real usage the template has behind it.

research-analysis
research-fact-checking
public-sector-nonprofit
nonprofit
grant evaluation
impact audit