Free tool
Call QA scorecard builder
Start from a template written for the calls your team actually makes, change what does not fit, and score a real call on it before you hand it to a reviewer. Everything stays in this browser.
No accountNo email neededDownload as CSV, PDF or a file you can reload
Start from
- Criteria
- 10
- Total weight
- 17
- Critical failures
- 3
- Sample score
- —
Opening
Discovery
Accuracy
Handling
Closing
Conduct
How the score is worked out
- A weighted average counts a criterion as many times as its weight; a simple average counts each once.
- A bonus criterion can add and can never take away — it sits outside the total a call is measured against.
- Not applicable leaves a criterion out of both sides, rather than scoring it zero for a question the call never raised.
- A critical failure is reported beside the score and never quietly rewrites it. A call can fail on one thing and still show what the rest of it was worth.
WhatsApp sales call
WhatsApp sales · weighted average · pass mark 80%
Agent ______________________ Call ______________________ Reviewer______________________ Date ____________
Opening
| Said who they are and why they are callingName, company, and the reason, in the first few seconds. | ___ / 5 | ×1 |
| Told the customer the call is recorded (critical)Said out loud, early, in a way the customer could object to. A recording nobody was told about is the one thing this scorecard treats as a failure of the call. | ___ / 5 | ×2 |
Discovery
| Asked what the customer actually needsQuestions before pitching. Did they learn anything they did not have from the chat? | ___ / 5 | ×2 |
Accuracy
| What was said about the product was correctPrice, availability, delivery, what is included. Checkable afterwards. | ___ / 5 | ×3 |
| Made no promise the company cannot keep (critical)No guaranteed outcomes, no dates nobody has confirmed. | ___ / 5 | ×2 |
Handling
| Handled the objection rather than talking past itNot applicable if the customer raised none. | ___ / 5 | ×2 |
Closing
| Agreed a specific next step, out loudA time, a document, a callback — something either side could hold the other to. | ___ / 5 | ×2 |
| Confirmed the next step back to the customer | ___ / 5 | ×1 |
| Followed up in the chat with what was agreed (bonus)Bonus: the written trail that stops a dispute a week later. | ___ / 5 | ×1 |
Conduct
| Courteous throughout, including under pressure (critical) | ___ / 5 | ×2 |
Critical failures are reported beside the score, not deducted from it. Built with the free scorecard builder at qualitify.net/tools/call-qa-scorecard
What makes a scorecard worth using
Weight the few things that decide the call
A scorecard where everything counts the same says that saying the wrong price matters as much as a slightly flat greeting. Pick two or three criteria that actually decide whether the call went well, weight those, and leave the rest at one. If every criterion is weighted, none of them is.
Keep critical failures rare and specific
A critical failure should describe something nobody would argue about afterwards — a recording the customer was never told about, a guaranteed return, a balance discussed with the wrong person. Three is plenty. If a reviewer is ticking one most weeks, it is not critical, it is coaching.
Ask for a reason when a mark is low
A low mark with no sentence beside it is a number an agent cannot do anything with, and the one they will dispute first. Set a threshold on the criteria where the mark is a judgement, and the scorecard asks for the reason while the reviewer still remembers the call.
Reviewing WhatsApp calls specifically
A call that began in a chat is different to judge. The customer has the earlier messages in front of them, so anything said on the call can be checked against what was written — which is why the accuracy criteria here are weighted heavily. The other difference is consent: on a channel people think of as personal, saying the call is recorded matters more, not less. Every template here makes that a criterion of its own.
Calibrate before you compare agents
Have two reviewers score the same call separately and compare. A gap of more than a point usually means a criterion is vague rather than that one of them is wrong, and it is far cheaper to find that now than after three months of scores nobody trusts.