Skip to main content

Lagging indicator

A metric that reports the consequence of a problem after it has occurred, such as a customer complaint count or a regulatory fine. Necessary for accountability but insufficient alone as a warning system, because it cannot prevent the harm it reports.

Defined in 4 GAGE programs, which carry 6 distinct definitions of it. The wording above is taught in AI Data Governance: The Data Chair.

How each discipline defines it

The same term does different work depending on who is using it. These are the definitions as each program teaches them, unedited.

AI Data Governance: The Data Chair

A metric that reports the consequence of a problem after it has occurred, such as a customer complaint count or a regulatory fine. Necessary for accountability but insufficient alone as a warning system, because it cannot prevent the harm it reports.

EU AI Act Implementation Expert

An outcome metric measured after events have occurred, such as the count of serious incidents reported under Article 73 or the remediation closure rate for Article 9 risk management actions.

Business AI Transformation

A metric that confirms an outcome after the fact (weeks to quarters), such as revenue attributed or retention. What an initiative is ultimately judged on, but too late to steer by alone.

AI Literacy & Professional Conduct

A later business outcome (such as revenue, tickets resolved, or customer satisfaction) that matters but is harder to attribute and arrives after the leading indicators move.

AI Literacy & Professional Conduct

A signal that moves only after a problem is well established (e.g. the headline accuracy metric); trustworthy but slow, so it should not be the only thing watched.

Where it is taught

The exact lessons this term appears in. The first 7 topics of every program are free with a free account.

Terms it appears with

Not an alphabetical neighbourhood: these are the terms taught in the same lessons, ranked by how often they appear together.