IT service desk quality assurance is the discipline most teams skip until something goes wrong — a frustrated user escalates to the CIO, a compliance audit flags poor documentation, or CSAT scores quietly slide for months before anyone notices. This guide explains what a practical QA process looks like, how to build one without drowning your team in admin, and which metrics and review habits actually move the needle.
Why Service Desk QA Is More Than Just Ticket Auditing
Most service desks treat quality as a reactive measure: check a ticket after a complaint, coach the agent, move on. That approach catches individual failures but never fixes the underlying system.
A structured IT service desk quality assurance process does something different. It samples tickets systematically, scores them against defined criteria, feeds results back into training and process improvement, and closes the loop so the same mistakes stop recurring.
The business case is straightforward. Poor ticket quality drives up reopen rates, damages user trust, inflates handle times, and creates compliance gaps. A consistent QA programme turns those costs into measurable savings and gives managers the evidence they need to justify headcount, tooling or training investment.
QA is also the connective tissue between other ITSM practices. It surfaces knowledge gaps that feed your knowledge management backlog, reveals categorisation errors that skew your reporting, and flags SLA risks before they become breaches.
What a Practical QA Framework Covers
A workable framework does not need to be complicated. It needs to be consistent. Most mature service desks build their QA process around four components.
Sampling Strategy
Random sampling is the baseline. Aim for a statistically meaningful slice — most experts recommend reviewing between five and ten percent of total ticket volume per agent per month. Supplement random sampling with targeted pulls: tickets that were reopened, tickets that breached SLA, tickets escalated by the user, and tickets closed in under two minutes (a common sign of premature closure).
Scorecard Design
A QA scorecard should measure what actually matters to service quality, not what is easiest to tick. Common criteria include:
- Correct categorisation and priority assignment
- Clear, jargon-free communication with the user
- Accurate diagnosis and resolution documented in the ticket
- Appropriate escalation or handover if the agent could not resolve
- SLA adherence and timely updates to the user
- Knowledge article linked or created where relevant
- Closure confirmed with the user before the ticket was closed
Weight the criteria by impact. Resolution accuracy and communication quality typically carry more weight than formatting.
Calibration Sessions
Scorecards only work if reviewers apply them consistently. Run a monthly calibration session where two or three reviewers independently score the same sample ticket, then compare results. Significant disagreement signals that a criterion is ambiguous and needs rewording. Calibration also prevents individual reviewers from drifting into personal preferences over time.
Feedback and Coaching Loop
Scores without follow-up are just numbers. Every reviewed ticket should generate a brief structured comment — what was done well, what could improve, and a specific action if a score fell below threshold. Aggregate scores monthly at agent and team level so coaching conversations are grounded in data rather than anecdote.
Building Your QA Scoring Rubric Step by Step
This is the section most guides skip. Here is a practical sequence for building a rubric from scratch.
- Start with your most common ticket types. A password reset has different quality markers than a complex application fault. Build separate rubrics or add ticket-type-specific criteria to a shared base rubric.
- Define what "good" looks like for each criterion before you define what "poor" looks like. It is easier to anchor scoring to a positive standard than to a list of failures.
- Use a simple numeric scale — three or five points per criterion is usually enough. More granularity adds subjectivity without adding insight.
- Pilot the rubric on a set of archived tickets before going live. Adjust any criterion where reviewers consistently disagree or where every ticket scores the same.
- Publish the rubric to agents before you start scoring. QA should not feel like surveillance. Agents who understand the criteria in advance perform better and accept feedback more readily.
- Review the rubric itself every quarter. As your service catalogue evolves, your quality criteria should evolve with them.
Connecting your rubric to the TIKTING service management platform means you can attach QA scores directly to ticket records, track trends over time, and surface low-scoring tickets automatically in team dashboards rather than managing spreadsheets.
Metrics That Tell You Whether QA Is Working
Running a QA process without measuring its impact is like running a service desk without SLAs. These are the metrics worth tracking.
QA Score Trend
Track average QA score per agent and per team over rolling three-month windows. A rising trend confirms that coaching is landing. A flat or declining trend after several coaching cycles is a signal that the problem is systemic — process, tooling or workload — rather than individual.
QA Score vs CSAT Correlation
If your QA scores are high but CSAT is low, your rubric is measuring the wrong things. If CSAT is high but QA scores are low, you may be over-engineering your criteria. The two metrics should move in the same direction over time.
Reopen Rate by Agent
Cross-referencing ticket reopen rates with QA scores is one of the fastest ways to validate whether your rubric captures real quality. Agents with low QA scores on resolution accuracy should also show higher reopen rates. If they do not, revisit that criterion.
Time to QA Review
If tickets sit unreviewed for more than two weeks, the feedback loses relevance for the agent. Track the average lag between ticket closure and QA review completion. Keep it under ten business days where possible.
Coaching Action Completion Rate
Log every coaching action generated by QA reviews. Track what percentage are completed within the agreed timeframe. Low completion rates indicate that QA is generating more admin than your team can absorb — a signal to simplify the process or redistribute reviewer workload.
Common QA Mistakes and How to Avoid Them
Even well-intentioned QA programmes fail. These are the patterns that derail them most often.
- Reviewing only escalated or failed tickets. This creates a skewed picture and demoralises agents who only receive feedback when something went wrong.
- Using QA scores in performance reviews before the process is calibrated. Agents will game an uncalibrated rubric, and you will lose trust in the data.
- Assigning QA reviews to the most senior agent rather than a dedicated reviewer or rotating panel. Senior agents are usually your highest-value resolvers — pulling them into review work has a real opportunity cost.
- Letting the review backlog build. A QA programme that runs three months behind is providing historical archaeology, not operational improvement.
- Failing to close the loop at the team level. Individual coaching is necessary but not sufficient. Monthly team QA reviews — where aggregate patterns are discussed openly — drive the systemic improvements that individual coaching cannot reach.
Pairing your QA process with endpoint visibility from Odysseus also helps in one specific way: many quality failures on hardware and software tickets stem from agents working with incomplete or stale asset data. When asset records are accurate and auto-discovered, agents spend less time guessing and more time resolving.
Frequently Asked Questions
What is IT service desk quality assurance?
IT service desk quality assurance is a structured process for sampling, scoring and reviewing support tickets against defined quality criteria. It identifies gaps in agent performance, process adherence and documentation standards, then feeds results into coaching and continual improvement cycles to raise the overall standard of service delivery over time.
How often should service desk tickets be reviewed for quality?
Most experts recommend reviewing tickets on a monthly cycle, sampling between five and ten percent of each agent's closed tickets. Supplement monthly reviews with triggered reviews for reopened tickets, SLA breaches and user-escalated cases. Quarterly team-level reviews should examine aggregate patterns rather than individual tickets.
Who should own the QA review process on a service desk?
Ownership typically sits with the service desk manager or a dedicated quality analyst role in larger teams. In smaller teams, a rotating panel of senior agents can share review duties, provided calibration sessions are held regularly to keep scoring consistent. QA ownership should be separate from day-to-day ticket resolution to avoid conflicts of interest.
What is the difference between QA and CSAT on a service desk?
QA measures internal process quality — whether agents followed the right steps, documented correctly and resolved accurately. CSAT measures the user's subjective experience. Both matter. QA without CSAT can optimise for compliance while missing what users actually care about. CSAT without QA makes it hard to diagnose why satisfaction is low or inconsistent.
How do you build a QA scorecard for a service desk?
Start by listing the criteria that define a well-handled ticket for your most common ticket types. Assign a numeric weight to each criterion based on its impact on resolution quality and user experience. Pilot the scorecard on archived tickets, calibrate with multiple reviewers, then publish it to agents before live scoring begins. Review and update the rubric quarterly.
Can QA scores be used in agent performance reviews?
QA scores can inform performance reviews once the process is calibrated and agents have had a reasonable period to adjust to the criteria — most teams wait at least one full quarter after launch. Scores should be one input among several, alongside CSAT, SLA adherence and ticket volume, rather than the sole determinant of performance ratings.
Key Takeaways
- IT service desk quality assurance requires systematic sampling, a calibrated scorecard, consistent feedback and a closed coaching loop — not just ad-hoc ticket audits.
- Build your rubric around what good looks like, pilot it on archived tickets, and publish it to agents before you start scoring.
- Track QA score trends, CSAT correlation, reopen rates and coaching completion rates to confirm the programme is driving real improvement.
- Avoid reviewing only failed tickets, using uncalibrated scores in appraisals, or letting the review backlog grow stale.
- Platforms like TIKTING can embed QA scores directly into ticket workflows, making the process easier to sustain at scale without separate spreadsheets or manual tracking.
