The documented origin and processing history of a record - where it came from, when it was collected, and how it was transformed.
Provenance is the trail attached to a piece of data: its source, the moment it was collected, and every transformation applied on the way to the consumer. It is what lets a research desk trust a value enough to size a position on it, and what lets a compliance team answer the unglamorous but essential question of where a dataset actually came from.
Provenance overlaps with lineage but is narrower and more record-level: lineage maps how data flows through a system, while provenance is the specific chain of custody for a given value. EdgeOrigin attaches provenance to every record so the output is auditable rather than just consumable - you can always trace a number back to the observation behind it, which matters as much for debugging a surprising signal as it does for an audit.
A live evaluation measures EdgeOrigin against your coverage requirements: decision-ready real-time data, delivery latency into your systems, and record-level provenance.