According to the white Enkata white paper Analytics Is Transforming Quality Assessment and Monitoring , typically the core of a company’s quality monitoring process involves listening to a very small number (sometimes less than five percent) of randomly selected call recordings and using a single form to score the agent’s performance. In some more mature companies, this process is supplemented with feedback manually gathered from completed customer survey forms. Given the nature of these processes, the results are:
• Limited in scope – Supervisors’ assessments of agent performance are based on a very small sample size. The evaluations also have a limited focus, looking largely at agent appraisals and training needs. The process does not take into account other sources of information about the agent’s performance such as contact records in a CRM system or data about how agents use desktop applications to resolve interactions, and may also not include what customers are saying on survey forms.
• Limited in validity – Our research shows that the vast majority of companies (about 80 percent) listen to only a small percentage of the calls they record, and nearly two-thirds leave it to the evaluator to select which calls are listened to. The outcomes are therefore of questionable validity as the findings will be influenced by the balance of good, average and bad calls chosen.
• Too subjective – Although companies have developed forms to help evaluators record their assessment, the process still relies on their interpretation of what they hear and their judgment as to how well it fits to the expected norm.
• Inflexible – Our research shows that nearly two-thirds of companies use a single form to carry out all their assessments. Only one-fifth have different forms for different calls, and even fewer (6 percent) vary the form depending on the agent being assessed. Evaluators are therefore limited to recording exceptions and comments from customer feedback surveys in an “other comments” section.
• One-off – The typical quality monitoring process is schedule-based, occurring at intervals specified by spreadsheets. Each assessment is treated in isolation and therefore does not provide the basis for a continuous agent improvement process. Also, assessments are likely to occur too late to impact the customer