Methodology & Sources
Banker outputs are built on a curated, source-backed coverage layer. The wider machine-extraction index sits behind it as a research and QA layer, never as a banker-ready fact by default.
80
core Environmental Services companies
135
broader published coverage
135
verified coverage names
65,662
extraction-grade records (secondary)
Coverage universe
The headline counts reconcile a curated universe to what has resolved to published data.
Main deliverables surface only clean user-facing counts. The configured-versus-published reconciliation lives here, by design, so the flagship pages stay free of competing numbers.
| Layer | Count | Basis |
|---|---|---|
| Verified coverage set | 135 | Hand-verified coverage roster (Environmental Services + cross-sector compounder benchmark) |
| - Core Environmental Services | 80 | Curated names in the six Environmental Services sub-sectors |
| - Compounder benchmark set | 55 | Cross-sector serial acquirers held for framework comparison |
| Broader published coverage | 135 | Configured names resolved to a published issuer index record |
| - Core ES resolved to published data | 80 | Core ES names with a resolved public-company data row backing every comp and cadence figure |
| Extraction-grade records (secondary) | 65,662 | Machine-extracted research rows for discovery, QA, and future curation; excluded from banker outputs by default |
All configured names resolved to published issuer records in this snapshot.
Two layers
Coverage-grade curation and extraction-grade research are kept strictly separate.
Coverage-grade curated layer
A coverage-grade record is part of the curated product layer: it has an assigned sub-sector, verified coverage status, an active or delisted label, and is eligible for banker outputs wherever the needed source-backed fields exist.
- All 135 configured names are verified; prioritization is a workflow view, not a data-quality distinction.
- Active and delisted labels are carried explicitly.
- Financial snapshots, M&A rollups, and named deals appear only where source-backed.
Extraction-grade research layer
Machine-extracted rows are useful for discovery, QA, audit, and future curation, but are not presented as banker-ready facts until reviewed and promoted.
- Accepted extraction status does not mean a record was promoted into coverage.
- Warning and quarantined rows require review before any promotion decision.
- The research layer is excluded from banker-ready outputs by default.
Source-backed evidence
The product favors SEC-backed evidence, labels conservatively, and shows N/A rather than guesses.
Values and labels
Missing values render as N/A, never null, undefined, NaN, or a false zero. Filing excerpts, ownership classifications, and transaction multiples are never fabricated.
- Disclosed consideration is not assumed to equal enterprise value.
- Transaction ratios use Disclosed consideration / revenue and / EBITDA unless enterprise value is confirmed.
- Evidence-strength labels indicate review confidence and source support, not investment merit.
Amounts and dates
Company-level acquisition amounts appear only where a source-backed figure passes sanity review; otherwise the value is withheld.
- Implausible or under-review amounts are shown as N/A, never as clean spend.
- Mixed SEC extraction values are never relabeled as acquisition spend.
- Future-dated rows are held out of flagship precedent tables unless clearly fiscal-year metadata.
Banker outputs
The Sector Book, Comps & Precedents, and Tear Sheet sit on the curated layer.
SE-Cluster is not investment advice, a deal-prediction engine, or a financing model. Ratios, capacity reads, and coverage rankings are source-backed research aids for coverage prioritization and diligence planning.