How records are rated and kept fresh
The discipline behind the hub: corroborated claims, links checked with real requests, explicit dates, and maintenance on a set cadence. The engine behind it is Crawl, the DAAC public-source engine.
The records on this hub, from the systems it profiles to the standards, regulation, initiatives, library and directory, are gathered and verified with one method: every material claim is corroborated, every external link checked, every figure dated. This page describes that method, notes what is already visible on the site, and explains how the hub is maintained.
Source ratings
Material claims are rated by the number, diversity and reliability of the independent sources behind them. A body publishing about itself counts as one source. Figures are dated.
| Rating | What it means |
|---|---|
| Strong | Three or more independent sources of different types agree. |
| Corroborated | Two independent sources, or one authoritative source plus corroboration. |
| Single source | One source only. Read with caution. |
| Contested | Credible sources disagree, or the only figure is a party’s own unaudited number. |
On the site today: the rating chip appears on every entry of the Standards, Regulation, Initiatives and Directory pages. On the Systems page the chip marks the system’s maturity; the source ratings for those records live in the hub’s data.
Link checks and dates
Every external link is verified with a real request to the URL, when a record is published and again on maintenance passes. A link’s status is never asserted without a check, and no date is ever fabricated.
A check means the link was verified and the entry read again on that date, not that every claim on the target page was re-fact-checked. On the site today these dates appear at section level (“Last reviewed July 2026” on Standards, for example) rather than as a per-record stamp; each record’s check date lives in the hub’s data.
Maintained, in the open
The hub is re-scanned on a set cadence. New records are added, links re-checked, dates revised. Candidates that fail verification are parked, not discarded, and retried on later passes, so coverage compounds. New entries and notable shifts are flagged in Latest.
Relationship to Delve
The hub carries the full public record of the World Bank and Pact Delve database ↗ and extends it. Every Delve resource, dataset, country profile, theme and project is featured here as a linked record in the library and registers, alongside sources Delve does not cover. Nothing is rehosted; each record links back to its origin.
Report a gap or a correction
If something is missing, wrong or out of date, tell us. Reports are logged when they arrive and handled in the next maintenance pass.
Fix a record, suggest a source
The Contribute page is the main path; for a one-off correction, a plain email works.
About Crawl
Crawl is a DAAC product: an agentic engine that scans a sector’s scattered public record, verifies findings by corroboration, organises them into a navigable hub, and maintains their freshness. This hub runs on it, alongside DAAC’s other reference hubs.
Could not substantiate
18 headline claims traced to source: 14 substantiated, 3 contested, 1 could not be substantiated. Figures we found in circulation but could not trace to a primary source at the standard this hub holds. They are listed here rather than published as fact.