What these numbers are
[AJ INPUT REQUIRED: Explain what benchmarks represent — operating ranges, typical performance by stage, what the data measures, etc.]
Where they come from
[AJ INPUT REQUIRED: Describe the engagement set (size, composition, timeframe), public cross-checks, sourcing rigor, caveats about representativeness]
What qualifies as an engagement
[AJ INPUT REQUIRED: Define the criteria for inclusion — company type, revenue floor/ceiling, geography, stage, any exclusions or special cases]
Date ranges covered
[AJ INPUT REQUIRED: Specify the time windows — trailing 12/24 months, reporting lag, how freshness is managed, when benchmarks are refreshed]
How metric definitions are normalized
[AJ INPUT REQUIRED: Explain formula standardization, handling of edge cases, unit conversions, assumptions about data quality or completeness]
Median, range, or judgment — which is which
[AJ INPUT REQUIRED: Clarify which summary statistic is reported for each metric class, when judgment calls override statistics, how bands are set (percentile vs. derived estimates)]
Review cadence
[AJ INPUT REQUIRED: State how often benchmarks are reviewed or updated, what triggers a refresh, how stability vs. currency is balanced]
Known limitations
[AJ INPUT REQUIRED: Name silent assumptions — survivor bias, geography skew, model applicability limits, what the data cannot answer, caveats for readers]