Feedback changes agent behaviour when it is specific, tied to a real call the agent remembers, delivered close to when the call happened, and followed up on to check whether it stuck. Generic monthly scorecards and vague reminders to be more empathetic rarely change anything because they give the agent nothing concrete to do differently. Coaching that works looks more like a short, focused conversation about one moment in one call, with a clear next action and a follow-up to see if it happened.
Contact centres invest heavily in quality assurance scoring and still see the same mistakes recur month after month. The scoring itself is rarely the problem. The gap is usually between the score and the actual coaching conversation: what gets said to the agent, how often, and whether anyone checks back in. Fixing that gap tends to matter more than fixing the scorecard.
Why Does Most Feedback Fail to Change Anything?
A few patterns show up repeatedly in contact centres that struggle with this. Feedback often arrives weeks after the call, by which point the agent has taken hundreds of other calls and cannot recall the specific moment being discussed. It is often delivered as a number on a scorecard rather than a conversation, which tells the agent they underperformed without explaining what to do instead. And it is rarely followed up, so there is no signal to the agent that it mattered enough for anyone to check again.
The Timing Problem
The closer coaching happens to the actual call, the more useful it is. An agent who gets feedback the same day, ideally referencing the specific call, can connect the feedback to a real memory of what they said and why. A month later, that same feedback lands as an abstract instruction with no context attached to it.
The Specificity Problem
Telling someone to improve your empathy is not something an agent can act on. Pointing out that when the customer said their delivery was three weeks late, you moved straight into the refund policy before acknowledging the frustration, try leading with an acknowledgement next time, is something an agent can actually practise. The difference between those two sentences is the difference between feedback that sits on a form and feedback that changes a habit.
How Should Coaching Sessions Actually Be Structured?
Structure matters more than length. A well-run ten-minute coaching session beats a rambling thirty-minute one because it forces both the supervisor and the agent to focus on one or two things rather than a laundry list.
- Start with the agent's own view, asking how they thought the call went before offering an assessment, which builds self-awareness rather than just compliance.
- Reference a specific moment, ideally by playing back a short clip of the actual call rather than describing it from memory or a scorecard note.
- Limit it to one or two focus areas, since an agent asked to fix five things at once usually fixes none of them.
- Agree on a concrete next action, something the agent can point to on their next relevant call as the thing they tried differently.
- Set a follow-up point, a specific call or date where the supervisor will check whether the change actually happened.
What Role Does Data Play in Better Coaching?
The single biggest lever for improving coaching quality is picking better calls to coach on. A supervisor with limited time who reviews calls at random will spend a good portion of that time on calls that were fine. Speech analytics tools can flag the calls most likely to contain a genuine teaching moment, such as ones with long silences, repeated customer questions, or a sentiment shift partway through, which lets a supervisor spend their limited coaching hours where they matter most.
Balancing Data With Judgement
Data should narrow down which calls to look at, not replace the supervisor's judgement about what happened in them. A flagged call still needs a human ear to understand context: sometimes a long silence is a struggling agent, and sometimes it is a legitimately complex case that took time to resolve properly. Treating every flag as a problem erodes trust with agents fast.
How Do You Coach Without Demoralising Agents?
Coaching that only ever points out what went wrong wears agents down and tends to increase attrition rather than lift quality. The centres that sustain good coaching cultures balance correction with genuine recognition, and they frame feedback as investment in the agent rather than punishment for a mistake.
Making Recognition Routine, Not Rare
If the only time an agent hears from a supervisor is when something went wrong, coaching starts to feel like a warning system rather than support. Building in regular, specific recognition for calls handled well, not just generic praise, keeps the relationship balanced and makes the corrective conversations easier to hear.
Connecting Coaching to Career Growth
Coaching lands better when agents can see where it leads. Linking feedback to a visible skills path, whether that is handling more complex queries, moving into a senior agent role, or qualifying for a different queue, gives agents a reason to actually want the feedback rather than just tolerate it. This is closely tied to why continuous training reduces attrition: agents who feel like they are developing are less likely to leave.
What Does Good Coaching Look Like at Scale?
A single dedicated supervisor coaching five agents can do this well through instinct and attention. A contact centre running hundreds of agents across multiple shifts and languages needs an actual system: a defined coaching cadence, a consistent format for sessions, calibration between supervisors so feedback does not vary wildly by team, and a way of tracking whether coached behaviours actually change over subsequent calls.
Calibration Between Supervisors
Without calibration, the same call handled the same way can get very different feedback depending on which supervisor happens to review it. Regular calibration sessions, where supervisors independently score the same sample calls and then compare notes, keep standards consistent and give agents confidence that feedback is fair rather than arbitrary.
Why Does This Matter More Than It Might Seem?
Coaching quality is one of the clearest differences between a contact centre that improves over time and one that plateaus. It is also one of the areas most visible in a contact centre's training programme to a client evaluating a partner, because it shows up directly in call quality, resolution rates, and how agents handle the client's specific product or service over time. A business weighing an in-house versus outsourced setup should ask a potential partner directly how coaching actually works day to day, not just what the scorecard measures, because the scorecard is only half the story.
Coaching Newer Versus Experienced Agents
Coaching that treats every agent the same, regardless of tenure, wastes time on both ends. New agents typically need more frequent, more structural coaching, focused on building the basic habits of good call handling before layering on nuance, and this early period is where the foundation from a structured training programme gets tested against real calls for the first time. Experienced agents already have those basic habits and benefit more from targeted coaching on specific gaps or on stretching into more complex work, since treating a long-tenured strong performer the same as a new hire risks turning coaching into a box-ticking exercise that stops feeling worthwhile to either side.
What Tools Actually Support Better Coaching?
Beyond speech analytics, a few other practical tools make coaching sessions more concrete and less reliant on a supervisor's memory or general impression of an agent's performance.
- Call recording with easy playback, so a supervisor can pull up the exact moment being discussed rather than describing it secondhand.
- A simple, consistent coaching log, tracking what was discussed, what action was agreed, and whether it was followed up on, which prevents the same feedback from being repeated without ever closing the loop.
- Peer coaching and call listening, letting strong performers share how they actually handle a difficult call type, which often lands better than the same advice delivered top-down from a supervisor.
None of these tools replace the actual conversation between supervisor and agent. They exist to make that conversation more specific and easier to follow up on, which is where most of the real value in coaching comes from.
Frequently Asked Questions
How often should agents receive one-on-one coaching?
Most well-run contact centres aim for at least one focused coaching session per agent per week, with shorter informal check-ins happening more often. The right frequency depends on tenure and performance, with newer or struggling agents typically needing more frequent contact than experienced, consistently strong performers.
Should coaching be based only on quality assurance scores?
Scores are a useful starting signal but should not be the entire basis for a coaching conversation. The most effective coaching combines the score with a specific call the agent can remember and reflect on, since a number alone rarely tells an agent what to actually do differently.
What is the biggest mistake supervisors make when coaching agents?
Trying to cover too many issues in a single session. When an agent leaves a coaching conversation with five things to fix, they typically retain and act on very few of them. Focusing each session on one or two specific, practisable changes produces far better results.
Does coaching quality differ between in-house and outsourced contact centres?
It depends entirely on how the specific centre operates rather than on the outsourcing model itself. A well-run outsourced partner with a structured coaching cadence and calibrated supervisors can coach more consistently than an under-resourced in-house team, and the reverse is also true.
How do you know if coaching is actually working?
The clearest signal is tracking whether the specific behaviour discussed in a coaching session actually shows up differently in the agent's next relevant calls, not just whether their overall score moves. Following up on individual coaching commitments, rather than only reviewing aggregate scores, is what closes the loop.
If you would like an honest, practical view on this for your own business, get in touch via Connect Centre Group's contact page.