Transparency
Where our sources disagree — and why we publish it
Most commodity platforms silently pick a "winner" when two authoritative sources publish different numbers for the same metric. We do the opposite — every inter-source disagreement is recorded, surfaced here, and left for readers to judge against each source's published methodology.
No silent winners
If FAO and World Bank disagree by > 5% on the same production figure, both values are kept. We don't paper over the gap with an editorial choice.
Reader judges methodology
Every source carries a methodology URL in its own API entry. Click through, compare how each side arrived at its number, and pick the one that fits your use case.
Corrections are logged
When a historical figure gets restated upstream, the change is archived (see migration 0009). Nothing silently changes underneath a cited chart.
Inter-source reconciliation ledger
Each row is one recorded disagreement between two of our upstream sources. The divergence threshold is 5% — below that, minor reporting lag is more likely than a genuine dispute.
This ledger is served from the public /api/v1/risk-intel/conflicts-public endpoint — free to redistribute under CC BY 4.0 with attribution. Every entry carries the source keys so you can walk back to either side of the disagreement via the metrics + export APIs.