Interaction Quality Score
Definition
Interaction Quality Score
Interaction quality score is the rating a reviewer gives one customer contact against a fixed scorecard for accuracy, process, and tone. It is quality judged from the inside, unlike the satisfaction ratings that customers themselves fill in after a call has ended.
The internal view catches what customers cannot. A caller has no way of knowing whether the agent skipped a compliance step or recorded the wrong disposition.
It is also the easier score to distort. Reviewers drift, scorecards get gamed — and without calibration two reviewers will grade the same contact very differently.
Key takeaways
- Interaction quality score rates one contact against weighted scorecard criteria.
- Reviewer calibration is what makes scores comparable between teams.
- Auto-fail items should sit outside the percentage, not inside it.
- Internal quality and customer satisfaction measure different things and will diverge.
How it works
An interaction quality score is produced by reviewing a recorded or logged contact against a weighted scorecard, awarding points per criterion, then expressing the total as a percentage of the points available.
The formula is: (points awarded ÷ points available) × 100.
Most scorecards group criteria into four families, and the weighting between them is a policy decision.
| Criterion family | Example item | Typical weight |
|---|---|---|
| Accuracy | Correct information given | Highest |
| Compliance | Required disclosure made | Auto-fail if missed |
| Process | Correct notes and disposition | Moderate |
| Communication | Tone, clarity, empathy | Moderate |
Compliance belongs outside the arithmetic. A missed disclosure should fail the contact outright rather than cost five points on a score of 100.
Calibration is the whole discipline. Running the same contact past every reviewer each month exposes drift — before it contaminates a quarter of reporting.
Quality management theory separates the two halves of this work. The American Society for Quality distinguishes quality assurance from quality control, drawing a line between the two.
Assurance provides confidence that requirements will be fulfilled, while control is the inspection side of actually fulfilling them.
Public services measure the customer-facing half separately. The U.S. federal customer experience program collects and publishes service-provider feedback on service quality.
Sample honestly. Reviewing three contacts per agent per month produces a score that says more about which contacts were chosen than about the agent.
Read it beside the customer view. Divergence between this score and customer satisfaction rating (CSAT) is a finding, not an error — one of the two instruments is measuring the wrong thing.
Never let the scorecard outlive the process. Criteria written for a product that changed two years ago reward behaviour nobody wants any more.
Examples
Scorecards differ by regulatory exposure, contact complexity, and how much of the review is automated. Five cases show how the score is built and used in practice.
Financial services weight compliance heaviest. Disclosure and identification failures are auto-fail items, and the remaining score barely matters when one is missed.
Healthcare lines score privacy handling first. Verifying identity before discussing records is binary, so it sits outside the point total entirely.
Technical support weights accuracy over tone. A polite wrong answer costs more than a blunt correct one, and the scorecard says so explicitly.
Retail teams weight communication highest. Because most contacts are simple, differentiation comes from how the conversation felt rather than what was said.
Outsourced providers calibrate jointly with the client. Monthly sessions where both sides score the same contacts keep the two organisations reading the same number the same way.
Related terms
Interaction quality score sits at the centre of contact-centre quality management. The terms below cover the review functions, the analysis tools, and the customer-side measures it should be read against.
- Call Quality Monitoring: the voice-channel review process that produces most scores.
- Quality Assurance: the function that owns the scorecard and the calibration.
- Quality Analyst: the role that reviews contacts and reports findings.
- Speech Analytics: the technology that widens sampling beyond manual review.
- Post-Call Survey: the customer-side instrument scored alongside this one.
- Customer Satisfaction Rating (CSAT): the external score internal quality is compared with.
- Coaching: the activity a quality score is supposed to feed.
FAQ
How is an interaction quality score calculated?
Award points per scorecard criterion, then divide points awarded by points available and multiply by 100.
What is a good interaction quality score?
Most operations target 85% or above, though the figure only means something once reviewers are calibrated.
Why should compliance items be auto-fail?
Because a missed disclosure is a breach rather than a deduction, and burying it inside a percentage hides it.
How many contacts should be reviewed per agent?
Enough that the sample is defensible. Three a month says more about selection than about performance.
Why does quality score diverge from satisfaction?
Because one measures process adherence and the other measures how the customer felt. Divergence is information.
How often should scorecards be revised?
At least annually, and immediately after any material process or product change.
Source partners running calibrated quality programmes for clients can compare models via Outsource Accelerator hubs.
Independent
