01 · Performance system

A weighted rating model that replaced subjective QA reviews

Before this existed, performance reviews on the team were a lead’s judgment call, applied inconsistently across testers at different levels. I designed a scoring system that replaces that judgment call with a formula anyone can audit.

The model scores every tester on two independent axes. Performance (60% of the score) is built entirely from objective data pulled out of Jira, QMetry, and attendance systems: bug quality and volume, test-case execution (volume, schedule accuracy, and cases-per-hour efficiency), and attendance. Potential (40% of the score) is four lead-assessed dimensions: impact and mentorship, communication, escalation handling, and process mindfulness. Each is scored 1 to 10 against a documented three-tier evidence rubric, so “exceeds expectations” means the same thing for every rater.

Neither axis uses one flat bar. Scores are role-adjusted: a QA Tester, a Senior QA Tester, and a Quality Analyst are held to different Low/Medium/High thresholds on the Performance axis, because a junior tester and a lead shouldn’t be judged against the same number. The two scores place every tester on a 9-box grid automatically, and the workbook includes a “how it works” reference tab so the model is replicable by someone else, not dependent on me to interpret it.