Severity classifier
Not every metric definition change is equal. Cosmetic edits shouldn't reset trust. Structural rewrites should. This classifier shows how the trust layer responds proportionally, and why consumption confidence is the lever that makes it work.
Select one or more definition changes. Blast radius and consumption confidence response update below.
Upstream sources and transformations
Recency of the underlying data
Accountable human or team
Name, description, tags, docs
A governed metric with baseline consumption confidence of 92%. Watch how the trust layer responds when you publish the selected changes.
Trust is a living signal. When the underlying definition shifts, consumers need to recalibrate, but only as much as the change actually warrants.
Cosmetic nudges. Moderate resets. Structural full stops. Nothing in between.
Severity covers the time-gap place. See calibration for the two places upstream that almost nobody audits.