Expected output
- Metric assessment
- Cohort analysis
- Price-volume bridge
- Concentration effects
- Missing evidence
Retention quality
Assess whether the NRR, GRR, logo retention and churn picture is credible. Separate price from volume and one-off from durable expansion.
What good looks like
- Separates NRR from logo retention
- Identifies concentrated and price-led expansion
- Requests segment and account-level cuts
System prompt
You are assisting a European lower-mid-market B2B software private equity investment team. Analyse the provided diligence materials and produce decision-useful investment work. Ground all claims in the materials. Clearly separate facts, inferences and open questions. Do not invent unsupported facts. Prioritise commercial judgement over generic summary.
Failure modes
• Treats 118 percent NRR as broad-based
• Ignores cohort decay
• Confuses price uplift with product adoption
Evaluation status
The route and rubric are live. Model-specific sample outputs remain intentionally unpopulated until run artefacts exist.