Loading
Reviewing a sample of handled tickets against agreed criteria — accuracy, tone, completeness, whether the process was followed.
Its value is twofold: coaching individuals, and finding process faults that make good work impossible. The second is usually the larger and the less used.
Random sampling finds the typical case. Also review the outliers on purpose: the escalations, the low CSAT scores, and the ones resolved suspiciously fast.
Those contain most of the learning. A purely random sample mostly confirms that routine tickets are handled routinely.
Enough for a pattern, with outliers deliberately included. Random-only sampling mostly confirms the routine.
No. Agents optimise for the rubric and avoid hard tickets.
The rubric is the problem. Calibrate before scoring anyone.
Half an hour with your own data usually saves reading three of these. The guides will still be here afterwards.