One corridor. Ninety days. A documented result.
A bounded corridor pilot with written scope, verification rules, publishable metrics, risks, and closeout.
Baseline
Document the selected geography, source availability, known limitations, initial recurring locations, and verification rules before measuring change.
Measures
Evaluate usable-case rate, duplicate-report relationships, recurrence after closure, repeat-site identification, verification turnaround, source provenance, and time-to-closure only where source data supports that measure.
Controls
Apply human review, privacy limits, documented uncertainty, access controls, and the commercial conflict firewall throughout the pilot.
Closeout
Publish the method, evidence limitations, results, and unresolved questions whether findings are favorable, mixed, inconclusive, or unfavorable.