Compare capture variants only after checking event coverage across native and browser paths

Before interpreting a capture experiment's retry-rate gap, audit whether each input path starts and finishes the same telemetry attempt. A metric can be internally consistent yet biased when browser-camera readiness initializes an older attempt tracker while native input initializes only a newer journey tracker. Join the two event families by flow after exposure and inspect failures absent from the headline metric. Count each flow once, separate voluntary retakes from validation failures, and preserve assignment after fallback. A failure event does not prove the user actually captured again; camera/session technical failures are another distinct measure.

A second provenance check is necessary: a shared completion callback can label manual captures as automatic. Verify the component's capture event before attributing wrong-subject images to auto-capture.

Applies when: comparing native camera, browser camera and automatic document capture using partially migrated analytics. Keywords: camera retry telemetry native browser experiment attribution missing events