Almost every supervisor eventually meets an agent who bristles at feedback. They argue the score, dismiss the example, or nod politely and change nothing. It is tempting to read this as a bad attitude, and the temptation leads many supervisors to escalate, apply pressure, and eventually lose a person who might have become a strong performer. In practice, resistance to feedback is usually a symptom with an identifiable cause, and addressing the cause works far better than pushing harder against the behavior.
Start by Understanding Why They Resist
Resistance rarely means an agent does not care. More often it reflects one of a few underlying causes:
- They do not trust the evaluation, because they have seen inconsistent scoring or feel judged on calls that were not representative
- They feel the feedback is personal rather than about specific behavior
- They believe the standard is unrealistic or unfair given their workload and systems
- They are anxious about their job security and respond defensively
- They disagree with the feedback and have not had a chance to say so
- They have had poor experiences with coaching before and expect more of the same
Each cause needs a different response, so the first coaching task is diagnosis. A short, genuine question such as “How did that call feel to you?” often reveals more than any amount of evidence.
Make the Evidence Hard to Argue With
A great deal of resistance is really a dispute about facts. When feedback rests on a supervisor’s impression, an agent can reasonably disagree. When it rests on specific moments from specific calls that the agent can see and hear for themselves, the argument loses most of its force.
This is where transparent scoring matters. If every score is linked to the exact point in the interaction that produced it, with the reasoning visible, the conversation shifts from “I do not agree with your score” to “Let us listen to this moment together.” Agents are also far more willing to accept feedback when they can see the same evidence as the supervisor, and when a fair process exists for challenging scores. Our post on handling QA score disputes goes deeper on keeping that process healthy.
Ask Before You Tell
Telling an agent what they did wrong invites defense. Asking them to evaluate their own performance invites reflection. A simple sequence works well:
- Play a short moment from a call and ask what they notice
- Ask what they were trying to achieve at that point
- Ask what they might do differently if it came up again
- Only then share your own observation, building on what they have already said
Agents often identify the problem themselves, and a conclusion they reach on their own is far more likely to stick than one they are handed. Research on feedback, including work published in Harvard Business Review, suggests people grow more from attention to what is working than from a catalog of what is wrong, which supports spending real time on strengths.
Separate the Behavior From the Person
Resistant agents often hear criticism as a statement about their worth. Careful language reduces that threat. Describe what happened, not what the agent is like. “On this call you moved to the solution before the customer finished explaining” is about a moment. “You do not listen” is about a person.
Keep the focus on one or two behaviors at a time. A long list of faults overwhelms and confirms the agent’s fear that nothing they do is good enough. One clear, achievable target gives them a way to succeed.
Acknowledge What Is Outside Their Control
Sometimes the agent is right. The system is slow, the knowledge base is out of date, or the call volume leaves no room to breathe. Brushing these aside destroys credibility. A supervisor who says “You are right that the system makes that step hard, and I am going to raise it” earns considerable goodwill, and then has standing to ask for change in the things the agent does control.
Data can help sort controllable from uncontrollable factors. If a pattern appears across many agents on the same interaction type, it is probably a process or tooling issue, not an individual failing, and treating it as the latter would be both unfair and ineffective.
Agree on Specific Next Steps and Follow Up
Coaching that ends without a plan is conversation, not coaching. Close each session by agreeing on one specific behavior to work on, how it will be measured, and when you will look at it again. Put a short follow-up in the calendar and keep it.
The follow-up is where trust is built. When an agent sees their effort acknowledged, that a measured improvement is recognized, and that the supervisor does what they said they would do, resistance usually softens. Gallup’s research on workplace engagement consistently points to regular, meaningful feedback and genuine interest in development as drivers of engagement.
Know When the Issue Is Bigger Than Coaching
Some situations genuinely require a different approach, such as repeated compliance breaches, behavior that harms customers, or sustained refusal to engage after fair support. In those cases, a clear and documented performance process is appropriate and fair to everyone. The key is to arrive there having given real coaching a proper chance, with evidence the agent has seen and a record of support offered.
Using Data to Make Coaching Easier
Resistant agents are far easier to coach when the preparation is strong. Supervisors who can see an individual’s trends across many calls, find specific representative moments quickly, and show improvement over time are working with credibility the agent can verify. Tools that surface this evidence automatically, such as conversation analytics and real-time guidance engagement data, save supervisors hours of preparation and put the focus on the conversation rather than the search for examples.
If you would like to see how ChorusCX helps supervisors coach with evidence agents can trust, book a demo with the team.