Call Center QA Scorecard: What to Include, With a Template

A call center QA scorecard should include five things: compliance items that are required on every call, the skills that drive the outcome (discovery, solution, objection handling, close), customer experience behaviors, measurable delivery checks like dead air and interruptions, and a short list of critical items that fail the call on their own. Each criterion needs a weight, a one sentence description of what a pass sounds like, and a rule for when it does not apply.

Below is how to build one, a template you can copy, and the mistakes that make scorecards hard to use. You can also grab the printable version and spreadsheet on our QA scorecard template page.

What is a call center QA scorecard?

A QA scorecard is the list of criteria a reviewer uses to grade a call, with a weight for each criterion and a definition of what passing looks like. It turns "that was a good call" into a score you can compare across agents, campaigns and weeks. A good scorecard is also a coaching tool: every line should point to something an agent can do differently on the next call.

The five sections every scorecard needs

1. Opening and compliance

These are the items a regulator, carrier or client expects on every call. They are usually yes or no.

• Agent states their name and company.
• Required recording or monitoring disclosure.
• Right party confirmed before details are discussed.
• Identity verified before account information is shared or changed.
• Any campaign specific disclosure, such as a debt collection disclosure or a Medicare disclaimer.

2. Discovery and understanding

This is where most sales and service calls are won or lost.

• Open questions about the customer's situation.
• Agent summarizes what they heard before proposing anything.
• Required qualifying details collected (for sales) or the issue restated correctly (for service).

3. Solution, accuracy and objection handling

• Information given matches policy, pricing and terms.
• No guarantees, pressure or claims the company cannot support.
• Objections acknowledged and answered, not ignored or argued with.
• Offer tied to what the customer said they need.

4. Close and next steps

• Agent asks for the sale, appointment, transfer or a specific follow up.
• Agent recaps what was agreed and what happens next.
• For service: issue resolved, or a clear owner and timeline given.

5. Delivery

Delivery covers how the agent sounds and handles the conversation. Some of it can be measured from the audio instead of judged by ear:

• No unexplained dead air (for example, silences of 5 seconds or more without "one moment while I check").
• Balanced talk time for the call type.
• Lets the customer finish without talking over them.
• Pace close to the customer's.
• Courteous, patient tone, including with a difficult customer.

Call center QA scorecard template

Here is a compact template for a mixed sales and service floor. Weights add up to 100. Items marked "Auto-fail" set the call score to zero when missed.

Change the criteria and weights to match each client's script. A collections line will add debt collection disclosures. A Medicare line will add the required disclaimers and scope of appointment checks. A support line may drop qualification and add empathy and ownership.

What the research says about what to score

Customer research backs up putting resolution, effort and tone on the scorecard, not just compliance:

• Resolution. SQM Group reports that every 1% improvement in first call resolution comes with a 1% improvement in customer satisfaction and a 1% reduction in operating costs. In its data, 5% of customers whose issue was resolved on the first contact said they intended to stop using the company, compared with 19% when the issue was left unresolved.
• Effort. Gartner reported in 2020 that customers who had a low-effort experience resolving their issue were 61% more likely to stay, and that a high-effort service experience dropped that probability to only 37%.
• Tone and empathy. In a 2024 OnePoll survey of 2,000 U.S. adults for TCN, 46% chose being pleasant to talk to or having empathy as the most important quality in a customer service agent.
• Experience as a whole. In Salesforce's 2023 State of the Connected Customer survey of more than 14,000 consumers and business buyers, 80% said the experience a company provides is as important as its products and services.

These are surveys and benchmarks, not proof that any one criterion moves your numbers. They are a good reason to give resolution and accuracy the most weight and to keep tone and effort on the card as real, scorable criteria.

How to weight a QA scorecard

Weights tell agents what matters. Put weight where revenue and risk are.

1. Start with the outcome. Ask what makes this call type succeed: a booked appointment, a resolved issue, a payment arrangement. Give the most weight to the behaviors that lead there.
2. Use auto-fails for hard compliance items. A missed required disclosure should not be balanced out by great rapport. Make it critical instead of giving it a huge weight.
3. Keep soft skills, but cap them. Tone and courtesy matter. They should not outweigh accuracy or the close.
4. Use partial credit sparingly. Partial works for criteria with several parts, like qualification. For yes or no items, it just creates disputes.

How to calculate the score

A simple, fair formula:

• Pass earns the full weight, partial earns half, fail earns zero.
• Criteria marked not applicable are removed from the total.
• Score = points earned divided by points available, times 100.
• If any auto-fail item is missed, the call is flagged as failed regardless of the score.

Worked example with hypothetical marks: a call has 90 points available after one 10 point criterion is marked not applicable. The agent earns 72. The score is 72 divided by 90, or 80. If that same call missed the recording disclosure, it is an auto-fail and goes to review.

How to write criteria that two reviewers score the same way

The biggest problem with most scorecards is wording. "Agent showed empathy" means something different to every reviewer.

• Describe the behavior, not the trait. "Agent acknowledged the customer's frustration in a specific way before moving to the fix" is scorable. "Agent was empathetic" is not.
• Say what counts as a pass in one sentence. If you need a paragraph, split the criterion.
• Define not applicable. "N/A if the customer raised no objections" prevents short calls from being punished.
• Quote the script where wording is required. If a disclosure must be read word for word, put the exact text in the criterion.
• Calibrate monthly. Have reviewers score the same few calls independently, compare, and fix any criterion they disagree on.

Common QA scorecard mistakes

• Too many criteria. Long forms get filled in quickly and carelessly. Aim for the smallest list that covers outcome and risk.
• Every item weighted the same. Equal weights tell agents that the greeting matters as much as the close.
• Compliance items scored as partial. Required disclosures are pass or fail.
• One scorecard for every campaign. Different lines have different rules.
• No evidence. A score without a quote or timestamp leads to arguments instead of coaching.

Scorecard by call type

See how these differ by industry on our pages for debt collection, insurance and Medicare and customer support.

How Sillana helps

Sillana scores calls against the scorecard you write, in the same format as this template: criteria with weights, pass descriptions, critical items and not applicable rules. Each mark comes with the quote from the transcript that supports it. Delivery items like dead air, talk ratio, pace and interruptions are measured from the audio rather than guessed. You can keep a different scorecard for each campaign, and every recorded call from a connected dialer gets scored, not a sample. Plans and limits are on the pricing page.

Key takeaways

• Include compliance, discovery, solution, close and delivery, plus a short list of auto-fails.
• Write each criterion as an observable behavior with a one sentence pass definition.
• Weight what drives the outcome and use auto-fails for hard compliance items.
• Remove not applicable criteria from the total so scores stay fair.
• Keep one scorecard per campaign and calibrate it regularly.

Questions people ask