How confident to be in any figure,
and how to ask for more.
How Narrows sources, tiers, and verifies the trade and routing data behind the Chokepoint Dependency Model and the Port Risk Index. Written for a buyer relying on this data, whether pricing a portfolio, feeding an ESG score, or financing a port asset.
Every figure carries a tier. The tier is never hidden.
This page explains how Narrows sources, tiers, and verifies the trade and routing data behind the Chokepoint Dependency Model and the Port Risk Index. It exists so that a buyer relying on this data can see how confident to be in any given figure, and how to ask for more.
Three tiers. Shown on every row, in the tool and in every export.
Every figure in the model carries one of three tiers. The tier reflects how the underlying value was sourced, not how important or how large it is.
Where each tier sits, by row and by value.
Counted by row, Estimated is the largest tier, concentrated in the long tail of smaller bilateral flows where official reporting is thinnest. Counted by the dollar value those rows represent, the picture reverses: Reported-tier rows carry the largest share of total modelled exposure. The flows with the most money behind them are disproportionately the ones with the strongest sourcing. We publish this breakdown both ways rather than a single confidence claim, so a buyer can judge how their own specific exposure sits, not take our word for the average.
Two checks run continuously, not only when something looks wrong.
Structured spot audits. A sample of country, commodity, and chokepoint combinations is checked against independent public sources on a fixed schedule, not only when an anomaly is reported. Each audit is logged internally: what was checked, what was found, and what was corrected.
Benchmark regression. A fixed set of known reference figures is checked after every model rebuild, so that a change made for one reason cannot silently move an unrelated figure without being caught.
Every correction, however small, is logged with a date, a description of what changed, the scope of countries or commodities affected, and the value impact. A public summary of this log is published as our data changelog, updated as corrections happen. The full internal log, including root cause detail and code level verification steps, is not published in full, for the same reason a pricing model's full workings are not published: the audit trail is the demonstration of rigor, not the underlying method. That fuller record is available to serious prospects under a non disclosure agreement, on request.
Any figure can be independently checked on request.
A specific country, commodity, or chokepoint figure can be independently checked on request. This produces a short verification note covering: the source of the figure, the tier it carries, the date it was last checked, and the result of a fresh spot check against an independent source if one has not been run recently for that combination.
This is offered rather than full access to the internal log because most of the value in a verification request is the answer, not the archive. A buyer who needs the full audit history for a specific regulatory or diligence purpose should raise that directly, and it will be considered case by case.
The current record lives on the changelog, not restated here.
As of end of August 2026, our data changelog is the current, dated record of every correction made to this model, updated as corrections happen. Recent activity includes four corrections closed between 13 and 18 August 2026, following seven made in early July.
The full audit log is available
under NDA for serious counterparties.
This page describes our process at a level sufficient to judge its rigor. The full internal audit log, including every check that found nothing wrong and the complete pre-July correction history, is available on request to qualified counterparties under a standard NDA.