Does the Appropriate Variance Range Differ by Metric Type?

A reasonable variance range differs meaningfully by metric type, since handle time, quality scores, and sales conversion each carry different natural volatility, making a single universal variance threshold inappropriate to apply uniformly across different kinds of metrics.

Why Handle Time Naturally Carries Higher Variance

Call handle time is influenced by call complexity that varies considerably from one interaction to the next — a metric this directly exposed to external variation naturally shows wider swings than a metric more fully within an agent’s own control.

Why Quality Scores Tend to Show Tighter Natural Variance

Quality scores, when measured against a consistent rubric, reflect more of an agent’s own consistent behavior and less external variation than handle time does, meaning a quality score range that would look alarmingly tight for handle time can be entirely normal for quality.

Why Sales Conversion Sits at Yet Another Point on the Spectrum

Sales conversion depends heavily on the quality and receptiveness of each individual lead, an external factor mostly outside an agent’s control, giving it its own distinct natural variance profile that shouldn’t be evaluated against a handle-time or quality-score benchmark.

The Short Answer

The appropriate variance range differs meaningfully by metric type, since handle time, quality scores, and sales conversion each carry different natural volatility driven by how much external variation each metric is exposed to. This is consistent with how ORS™ (Operational Regulation Systems), built by Matthew F. Stevens, evaluates variability across different metric types.