Many companies implement internal Qualy Assessment (QA) programs, using the data generated as part of a front-line
agent’s performance review. In some cases, they reward agents with incentives based on QA
scores. We have, however, found it quite common for companies to design their scoring tool and set up the program with little regard for statistical validity, making the resulting data suspect.
Common statistical problems include sample sizes that are inadequate to provide reliable results at
the individual agent level or scoring tools that could never generate objective, statistically valid data. Let me ask you a reasonable question: If an unhappy employee (or former
employee) chose to take issue with your QA process, would it stand the test? It is
imperative for any QA program to give consideration to their scoring
methodology and sampling approach. Only then can you have confidence in the data
gleaned from your analyst team.