Call Center QA Scorecard for Pay-Per-Call: What to Score and How

In pay-per-call, quality is not a soft metric. Every call you send is reviewed by someone on the buyer's side, and the calls they do not like come back as returns, lower caps or a quieter inbox. A QA scorecard is how your own team catches those calls first.

The goal is not paperwork. The goal is a short, consistent way to listen to a call, decide whether it was good, and tell the agent exactly what to do differently tomorrow. Below is the structure we use as a starting point. Adapt it to your verticals, your buyers' specs and your compliance process.

Start with what the buyer pays for

Before you write a single scoring line, pull the buyer's spec for each campaign and answer three questions:

1. What makes a call billable? Usually a minimum duration plus a qualified caller in an allowed state, ZIP code or age range.
2. What makes a call returnable? Wrong state, already a customer, no consent, wrong product, abusive caller, duplicate.
3. What does the buyer's agent need at the handoff? Name, a confirmed reason for calling, and a warm, clean transfer.

Your scorecard should measure exactly these things. If a line on the form does not connect to billing, returns or compliance, consider removing it. Short forms get used. Long forms get skipped.

The five sections of a pay-per-call QA scorecard

1. Opening and disclosures

• Agent greets the caller with the approved opening, word for word where the script requires it.
• Required disclosures are read as written, not paraphrased.
• Call recording notice is given where your process requires it.
• Consent is confirmed in the wording your compliance counsel has approved.

Treat this section as pass or fail. A call with a missing disclosure is a failed call, no matter how well the rest went.

2. Qualification

• Every qualifying question on the buyer's spec is asked, in the order the buyer expects.
• Answers are listened to, not assumed. The agent confirms state, ZIP code or eligibility when the caller is unclear.
• Callers who do not qualify are handled politely and not transferred.

This is where most returns are born. A caller from a state the buyer does not cover, transferred because the agent rushed, is a call you pay for twice: once in agent time and once in the return.

3. Conversation quality

• The agent speaks clearly, at a steady pace, without reading like a robot.
• Objections are handled with the approved responses, not improvised promises.
• No pressure, no exaggeration, no claims the buyer has not approved.
• The caller is never talked over or rushed.

Score this section on a simple scale (for example, 0 to 2 per line). It is the most subjective part of the form, so calibrate it often.

4. Transfer and handoff

• The agent explains the transfer before making it.
• The transfer is warm where the campaign requires it: the agent introduces the caller to the buyer's agent with the key details.
• No dead air, no hold beyond what the process allows, no dropped transfer.
• If the buyer does not answer, the agent follows the fallback rule instead of leaving the caller waiting.

5. Wrap-up and records

• Disposition is set correctly in the dialer.
• Any request not to be called again is added to the internal do-not-call list during or immediately after the call.
• Notes are entered where the campaign requires them.

How to weight the score

Keep the math simple enough that an agent can follow it.

• Compliance lines are gates, not points. A missed disclosure or consent line fails the call regardless of the total.
• Qualification carries the most weight among the scored sections, because it drives billing and returns.
• Conversation and transfer come next.
• Wrap-up is weighted lightly, but a missed do-not-call entry is treated as a compliance failure.

Write the weighting on the form itself. Agents should never have to guess how they were scored.

How many calls to review

There is no single right number, but there are good rules:

1. Review a set number of calls per agent per week, chosen at random, not only the calls that were flagged.
2. Add every returned or disputed call to the review queue, so you learn why it came back.
3. Review more calls for new agents, agents on a new campaign, and any agent whose returns are rising.
4. Include short calls that ended just before the billable duration. They often reveal a script or routing problem rather than an agent problem.

Listen to full recordings, not just the first minute. Many problems show up at the transfer.

Calibration: make sure everyone scores the same way

A scorecard is only fair if two reviewers would give the same call the same score. Once a week:

1. Pick three or four recorded calls.
2. Have every QA reviewer and floor manager score them separately.
3. Compare the scores line by line and agree on the right answer.
4. Write down any rule you clarified, and add it to the scorecard guide.

If you can, invite a buyer to join a calibration session once in a while. Hearing how the buyer scores a call is the fastest way to close the gap between your view of quality and theirs.

Turning scores into coaching

Scores that sit in a spreadsheet change nothing. Close the loop every day:

• Same day feedback. Share the result with the agent before the next shift, privately.
• One point at a time. Pick the single most important fix, play the moment from the recording, and model the better version.
• Recognise good calls. Play a strong call to the team in the pre-shift briefing. Agents learn faster from a good example than from a list of mistakes.
• Track the trend, not the single call. Look at each agent's results over two or three weeks before making decisions about campaigns or seats.

Our daily floor checklist builds live monitoring and same-day feedback into every shift, so QA is part of the routine rather than an extra task.

Use QA data to fix the system, not just the agent

When many agents fail the same line, the problem is rarely the agents. Look upstream:

• Script problems. A qualifying question that callers keep misunderstanding needs rewording, with buyer and compliance sign-off.
• Routing problems. Callers from states the buyer does not cover point to a filter or tracking setting, not an agent.
• Source problems. If calls from one publisher or one list consistently score low, raise it with that source and review the list hygiene behind it.
• Training problems. A new campaign with low scores across the board usually means the briefing was too short.

Bring a short QA summary to your weekly review: pass rate by campaign, top three failed lines, and what you changed because of them. If you are still building your operation, the first 30 days guide shows where QA fits into the launch plan.

A one-page starting template

Print this and adapt it:

1. Opening and disclosures: pass or fail.
2. Consent confirmed in approved wording: pass or fail.
3. All buyer qualifying questions asked, in order: scored.
4. Non-qualified callers handled and not transferred: scored.
5. Clear, calm, no unapproved claims: scored.
6. Transfer explained, warm where required, no dead air: scored.
7. Correct disposition and do-not-call requests recorded: scored, with do-not-call misses treated as a fail.
8. One coaching point for the agent: written.

Keep the form on one page, review it every month, and have your compliance counsel check the disclosure and consent lines whenever a script or campaign changes. A floor that scores its own calls honestly sends better calls, and better calls are what keep buyers coming back.