Empathy scoring is a QA practice that specifically rates whether an agent acknowledged a customer's emotions and responded to them appropriately, separate from whether the agent solved the problem correctly.

‍

‍Why it matters

A technically correct answer delivered coldly can leave a frustrated customer feeling worse, not better, and standard accuracy-focused QA doesn't catch that at all, since it's built to check whether the right information was given, not how it landed emotionally. Empathy scoring closes that specific gap, especially for complaints, cancellations, or an obviously upset customer.

‍

‍How it's typically scored

A basic rubric checks a few concrete things, did the agent acknowledge the customer's frustration in words, did they avoid sounding scripted, did their tone match the seriousness of the situation. Because this is more subjective than checking whether a policy was followed, empathy scoring benefits more than most QA categories from calibration sessions, since without them, reviewers can score the identical response very differently.