Data can be bought anywhere. The hard part is knowing whether it is true — where it came from, when it was last refreshed, whether it agrees with other sources, and who answers for it if it turns out to be wrong. That is the work we do, and that is what separates buying here from buying straight from a source you do not know.
No country is a silent default. Region rules, identity formats, prohibited categories, currency, and tax all come from a Jurisdiction Pack — configuration, not code. Adding a country means adding a pack, not rewriting the application.
The scoring formula is published. The margin range is published. Endpoint SLAs sit next to uptime we actually measured. If we cannot stand behind a number, we do not print it.
Data with nothing to compare against is marked “not cross-verified yet” — not given a zero that reads like a judgement. That difference matters to anyone deciding on the strength of a score.
Who accessed what, when, from which jurisdiction — recorded append-only from the very first transaction. A log installed after the trouble starts cannot fill in the past.
The legal entity is verified against an official registry, and proof of resale rights is required. This is a hard gate, not a formality.
Structure, anchors, triangulation, anomalies, period-to-period consistency, and dead-data detection. The result becomes a four-dimension score with readable reasons.
Schema, types, units, and region codes are normalised; the original columns are kept alongside. Every release carries a semantic version and a changelog.
Before each transaction the system checks whether the buyer’s jurisdiction may receive that data. If not, the product stays visible and the buy button is disabled with the reason shown.
Each pack carries its own versioned master regions, anchor data, validation rules, prohibited categories, currency, and tax rate.