QA, Coaching and Compliance Are Different Functions
Combining them into one process is the most common structural mistake in contact centre quality, and it makes the coaching adversarial and the compliance unreliable.
One evaluator, one form, one score, serving three purposes with different audiences and different consequences. The conflict is structural.
The three, separated
Compliance monitoring. Verifies that regulated or mandatory steps occurred. Audience: risk, legal, the regulator. Consequence: remediation, sometimes discipline. Needs completeness and auditability.
Quality assurance. Measures how well work is done, in aggregate, to find where the operation needs to improve. Audience: operations management. Consequence: process change, training programmes. Needs consistency and enough sample to be meaningful.
Coaching. Helps an individual agent get better. Audience: the agent. Consequence: development. Needs specificity, trust and frequency.
Why combining them fails
Coaching becomes adversarial. If the same conversation determines a score that affects the agent's standing, the agent defends rather than listens. Every coaching session is a negotiation about the score.
Compliance becomes negotiable. A missed disclosure weighted at five points, offset by good rapport, is not a compliance finding. Regulators do not accept a percentage.
QA becomes individual. Aggregate insight about a broken process gets delivered as feedback to whichever agent was scored, when the finding is about the knowledge base or the system.
The evaluator's role is incoherent. They are simultaneously an auditor, an analyst and a coach, and those require different relationships with the agent.
Separating them in practice
Compliance: automated presence checks across all interactions, with human confirmation of any failure. Reported as a list of exceptions, not a score. Owned by risk or compliance.
Quality assurance: aggregate analysis across teams and processes, using both analytics and a human sample. Reported as findings about the operation. Owned by operations.
Coaching: the supervisor, using specific interactions as material, separated from any scoring conversation. Owned by the line manager.
The same evaluation data can feed all three. What differs is who receives it, in what form, and what happens next.
The score-linked-to-pay question
Where QA scores affect compensation, the coaching function is effectively destroyed, because no agent treats a conversation about their pay as development.
If scores must feed pay, run coaching as a separate conversation with a separate person, and make it explicit that it does not affect the score.
Better, where possible: link pay to outcomes — resolution, customer feedback, attendance — rather than to a QA percentage with the sampling problems described elsewhere.
The QA score is a poor pay instrument because its uncertainty is large, its weights are arbitrary and its evaluators disagree. Using it that way invites disputes that are statistically justified.
What each needs from the other
They are separate and not independent.
Compliance findings feed coaching, for the individual, and QA, for the pattern.
QA findings feed process improvement and identify what coaching should focus on.
Coaching outcomes feed QA, by showing whether the identified problems are being fixed.
The connection should be through the data, not through a single meeting where all three happen at once and none of them works.
The organisational test
Who owns the QA function?
Reporting into operations produces pressure to show good scores. Reporting into compliance produces a policing posture. Reporting independently, with results going to both, is the arrangement that survives, and it is uncommon because it costs a reporting line nobody wants to create.
Who reports to whom
The reporting line determines the programme's behaviour more than its methodology.
QA reporting into operations produces pressure toward favourable scores, because the person receiving the report is accountable for the number.
QA reporting into compliance produces a policing posture and weak engagement with process improvement.
QA reporting independently, with results going to both operations and compliance, is the arrangement that survives scrutiny.
It is uncommon because it requires a reporting line nobody wants to create and a headcount nobody wants to own.
The workable compromise where independence is not achievable: an external or cross-site calibration requirement, a dispute route that escalates outside the operational line, and compliance receiving the failure data directly rather than through operations.
Whatever the structure, write down who can change the scorecard, because that is where pressure lands first.