QA evaluation sheets
Definition
QA evaluation sheets
QA evaluation sheets are the weighted scorecards a quality analyst uses to grade a call, chat or email against fixed criteria, so that a subjective judgement becomes a repeatable number that every analyst on the floor, and the client reading it, can defend.
Most BPO floors run one sheet per channel and one per programme, never a single master form. A voice form for a telco reads nothing like a chat form for an e-commerce account, but the bones stay the same.
The form is the unit of measurement — coaching plans, bonus calculations, calibration sessions and quarterly client reviews all trace back to the score written on it. Get the form wrong and everything downstream inherits the error.
Practitioner libraries such as ICMI’s contact-center resource library treat scorecard design as craft, not paperwork. Three choices carry most of the weight: how sections are weighted, how fatal items behave, and how tightly analysts are calibrated.
Key takeaways
- A QA evaluation sheet converts one interaction into a weighted, comparable score.
- Most BPOs run one sheet per channel and one per programme, not a single master form.
- Contact centres typically grade 4–8 interactions per agent per month for a usable sample.
- Fatal errors auto-zero the sheet however well the other line items scored.
- Analyst variance under about 5% is a common operator target, not a published standard.
How it works
A scorecard moves through a fixed loop, from form design to version control. Most contact centres grade four to eight interactions per agent each month, enough to be statistically useful without swallowing the QA team’s week.
| Step | What happens | Who owns it |
|---|---|---|
| 1. Form design | Agree sections, weights and fatal items with the client before scoring starts | QA manager |
| 2. Sample selection | Pull calls at random or by trigger: escalations, repeat callers, low CSAT | QA lead |
| 3. Score the form | Mark each line item, add comments, calculate the weighted total | QA analyst |
| 4. Calibration | Several analysts score one interaction and reconcile the gaps | QA + ops |
| 5. Agent feedback | Walk the agent through the score and agree one coaching focus | Team leader |
| 6. Trend reporting | Aggregate scores feed weekly dashboards and client reports | QA manager |
| 7. Version control | Re-weight and re-issue the form when policy changes, then archive the old one | QA manager |
Weighting is where the form does its real work. A typical voice scorecard splits into three scored buckets plus items that sit outside the maths, and the split tells an agent what the programme really cares about.
| Section | Typical weight | What it covers |
|---|---|---|
| Soft skills | 25–35% | Greeting, tone, empathy, active listening |
| Process compliance | 30–40% | Scripts, disclosures, system notes, verification |
| Resolution and accuracy | the balance | Correct answer, correct action, no repeat contact |
| Fatal items | auto-zero | Compliance, data protection, conduct breaches |
Fatal errors sit outside the weighting. One fatal zeroes the sheet however well the agent scored elsewhere. Typical fatals are a missed compliance disclosure, a data-protection breach, abusive language, or giving wrong financial or medical information.
Calibration stops the form becoming one analyst’s opinion — analysts grade the same recording blind, compare it line by line, then argue out the gaps until each criterion means one thing to everyone scoring it.
Most floors chase analyst-to-analyst variance of under about 5%. Treat that as a working convention, not a rule handed down: no standards body publishes a calibration variance threshold, and the number travels by word of mouth between operations teams.
The COPC Customer Experience Standard, now at Release 8.0, is the framework most enterprise BPOs design their forms against. Read it closely, though, and it prescribes far less than people assume.
COPC sets out a performance management system and tells operators to choose the measures that fit their own business. The design burden stays with you, which is why two BPOs serving the same industry ship very different forms.
Examples
Five operators show how the same weighted-criteria bones flex across channels, programmes and geographies. Voice forms for regulated accounts weight compliance hardest, chat forms grade written etiquette, and outbound programmes carry fatals a service desk never needs.
Concentrix (Manila, ongoing). Concentrix runs programme-specific scorecards across its Philippine sites for banking and healthcare clients. Regulated accounts weight resolution and compliance hardest, and weekly calibration keeps two analysts scoring one call alike.
TP, formerly Teleperformance (2024). TP layered speech analytics onto manual QA sheets under its TAP framework. Auto-scored items such as silence, talk-over and sentiment feed the same sheet an analyst grades, so the form blends machine marks with human ones.
Scale is the reason. TP employed over 485,000 people across nearly 100 countries at the end of 2025, and its 2024 integrated report puts interaction volume above one billion a year.
Foundever, formerly Sitel Group. Foundever publishes scorecard guidance that treats the form as a coaching artefact first and a compliance tool second. Worked “what good looks like” examples sit beside each line item, so agents can self-review before the next call.
TaskUs (as of 30 June 2025). TaskUs ran roughly 60,400 people across 30 locations in 13 countries. At that spread, version control stops being admin — a form edited in Manila and never re-issued in Bogotá scores identical work two different ways.
A Cebu BPO on outbound sales (2025). A 20-line voice scorecard carried three fatals: no Do Not Call check, no recording disclosure, no payment-card masking. Analysts graded six calls per agent a month, and any fatal pulled the agent off the queue.
Related terms
The cluster around this term splits three ways: the function that owns the form, the session that keeps it honest, and the metrics that sit beside the score without living inside it. The scorecard measures behaviour; the terms below measure outcomes.
- Quality Assurance: the function that owns the scorecard and the analysts scoring it.
- Call Calibration: the session where analysts grade one call and reconcile their scores.
- Average Handle Time: a productivity metric that sits alongside QA scores, not inside them.
- First Call Resolution: an outcome metric many scorecards weight heavily.
- Customer Satisfaction Score: the customer-side companion to the analyst-side QA score.
- Net Promoter Score: a loyalty metric clients often pair with QA scores in quarterly reviews.
- Service Level Agreement: the contract that defines what passing QA means for a client.
FAQ
What is a QA evaluation sheet used for?
It grades an interaction against fixed, weighted criteria so you can coach consistently, calibrate analysts and report to a client. The form turns a subjective call into a number two people can argue about from the same evidence.
How many calls should be evaluated per agent per month?
Most contact centres land between four and eight. Below four, the sample says more about which calls you picked than about the agent. Above eight, the QA team burns hours without changing the coaching conversation.
Who fills out the QA evaluation sheet?
A dedicated QA analyst, not the agent’s own team leader. The analyst owns scoring objectivity and the team leader owns the coaching that follows, which stops the score becoming a bargaining chip.
What’s the difference between QA scores and CSAT?
QA scores are internal: an analyst grading the agent against your standards. CSAT is external: the customer rating their own experience. The two correlate imperfectly, which is why mature operations track both and reconcile the gap monthly.
What counts as a fatal error on a scorecard?
A fatal auto-zeros the sheet however the rest of it scored. Typical fatals are a missed disclosure, a data breach, abusive language or wrong financial advice. Keep the list short — a dozen fatals stop telling a bad call from a catastrophic one.
Can QA evaluation sheets work for chat and email too?
Yes, with different line items: voice forms grade tone and active listening, chat forms grade typing accuracy, and email forms grade structure and resolution completeness.
Explore more outsourcing terms and provider guidance at Outsource Accelerator.
Independent
