A day in the life of a quality analyst

Quality · A day in the life · ~7 minute read

The QM analyst job is judgement-heavy in ways the scoring sheet doesn’t admit. Every score is contestable, every theme is partial, every calibration session is a small political negotiation. Here’s what the role actually looks like.

8:30am — the scoring batch

Open the QM tool. Today’s sample: 30 voice calls, 15 chats, 5 written complaints. Score each against the scorecard. The honest QM analyst takes 8–12 minutes per call — less and you’re ticking boxes; more and you’re over-engineering judgement that won’t hold up in calibration.

A discipline: note your confidence level on each judgement call. "Score 7/10 on empathy — high confidence" vs "Score 7/10 on judgement — low confidence, borderline call." In calibration, the low-confidence judgements are where you learn most.

11:00am — the calibration session

Three QM analysts plus the QM lead, plus a TL, plus the head of customer experience. Six people, four calls. The intent: agree the score. The reality: surface where the scorecard is ambiguous, where it’s being applied differently, where it’s being weaponised against a particular team.

The honest analyst defends their score with reasoning, accepts when reasoning was weaker than they thought, and pushes back when the room pressure is to soften a score that shouldn’t be softened. Politics happen in calibration; the disciplined response is to name them rather than absorb them. "I think we’re softening this because the agent is well-liked, not because the behaviour was actually fine. Let’s be explicit about which."

13:00pm — coaching prep

Pulling together feedback for the TL who’s coaching an agent tomorrow. The QM analyst’s job: not to coach (that’s the TL’s) but to give the TL the evidence, the patterns, and the specific calls to use as coaching examples.

A discipline: positive examples as well as negative. If you give the TL only the bad calls, the coaching becomes about deficit. If you give them the good ones too, the coaching becomes about replication. The good QM analyst hands over both, with the calibrated reasoning for each.

15:00 — the meeting where QM is asked to flatter

A team manager has an underperforming agent. The team manager wants QM to "be supportive in next month’s scoring." Not explicitly — just a vibe, a few suggestions, a hint that calibration could go a particular way.

The disciplined QM analyst notices and refuses calmly. "We’ll score what we see. If the trajectory is positive, that’ll show. If it’s not, you should know that before you stand in front of the agent." The team manager who pressed has now learned this analyst doesn’t play that game. Reputation compounds; the analyst who folds once is asked to fold again.

16:30 — the theme work

The last hour. Look across the week’s scoring for themes the operation should know about. AHT drift on a specific contact type. Vulnerability cues being missed. A particular product driving 60% of complaints.

Theme work is what makes QM useful at the operation level. Most QM functions deliver agent-level scoring and stop there. The ones that turn the scoring into themes the planner, the product team, and the operations leader can act on are the ones that earn their seat at the table.

The QM analyst’s real value

Score honestly. Defend reasoning. Refuse pressure calmly. Hand TLs evidence that helps them coach. Surface themes the operation can act on. The QM analyst who does these things becomes a respected operational voice; the one who scores high volume and lands no themes becomes a cost the operation eventually questions.