“Enhance process quality” should be measured by a KPI that captures defects or errors in the process output. “Error rate (%)” directly reflects quality performance by quantifying the proportion of transactions/outputs that contain errors, fail checks, or require rework. Option A (training attendance) is a leading/input measure—useful as a driver but not proof that quality improved. Option B is written like a target statement/initiative-style goal rather than a KPI definition; it mixes a desired level with a deadline instead of defining the metric itself. Option D (time to process a transaction) measures speed/efficiency , not quality; improving speed can even harm quality if not balanced. A common measurement challenge for error rate is consistent defect definition and detection (what counts as an error, where it’s recorded, and whether audits are consistent). Activation best practice includes clear defect taxonomy, sampling rules (100% check vs audit), and a balanced dashboard pairing error rate with cycle time so teams improve quality without creating bottlenecks or encouraging underreporting.