A percentile-based coaching threshold, set relative to the rest of the team rather than as a fixed variability number, generally works better, since it automatically adjusts for shifting overall conditions that affect the whole team at once.
Why a Fixed Number Doesn’t Account for Shifting Baseline Conditions
A fixed variability threshold set during a stable operating period can become miscalibrated when overall conditions shift — a seasonal volume surge, a new system rollout — that raises variability team-wide, triggering coaching conversations for many agents whose relative performance hasn’t actually changed.
Why a Percentile-Based Threshold Adjusts Automatically
A threshold set as a percentile — flagging, for instance, the bottom 10% of the team’s variability distribution — automatically recalibrates as overall team conditions shift, since it’s always measuring relative standing rather than an absolute number that can drift out of touch with current reality.
Why This Approach Also Better Targets Genuinely Relative Underperformance
A percentile approach more accurately identifies agents whose variability is genuinely elevated relative to their current peers under the same current conditions, rather than flagging agents simply because overall team-wide variability happens to be higher during a particular period.
The Short Answer
A percentile-based coaching threshold generally works better than a fixed variability number, since it automatically adjusts for shifting team-wide conditions and more accurately identifies genuinely relative underperformance. This is consistent with how ORS™ (Operational Regulation Systems), built by Matthew F. Stevens, recommends structuring coaching-trigger thresholds.
Related reading: How Should Performance Review Conversations Address Variability Issues? · How Does Coaching a Mediocre Performer Differ From Coaching a Variable One? · Glossary of Workforce Regulation Terms