Creating Audit Scorecards

Audit scorecards define how your team scores quality during reviews. They turn quality from a loose opinion into a repeatable standard.

Use this guide when you want to set clearer expectations for conversation and ticket review.

What an audit scorecard is

An audit scorecard is an ordered set of criteria your team uses during quality reviews.

A strong scorecard helps your team answer questions like:

Was the answer accurate?
Was the tone appropriate?
Did the issue reach a real resolution?
Did the handoff or follow-up meet expectations?

Conversation scorecards vs ticket scorecards

Use a conversation scorecard when you want to score the customer interaction itself.

Use a ticket scorecard when you want to score the quality of follow-up work, investigation, or completion after the conversation moves into a longer-running process.

Some teams use both because the conversation and the follow-up work require different standards.

What makes a good scorecard

A good scorecard is:

Clear enough that two reviewers would score similarly
Short enough to use consistently
Focused on meaningful quality signals

Start with a small number of criteria that matter most. You can expand later once the team is using the scorecard consistently.

Create a scorecard

Open Settings -> Workspace -> Audit -> Criteria and create the reusable criteria your team needs.
Open Audit -> Scorecards and create a new scorecard.
Choose whether it applies to conversations or tickets.
Add the criteria your reviewers should score.
Put the criteria in the order reviewers should work through them.
Choose the scorecard from Audit metric when you start the audit.

Every new audit requires a scorecard. The order matters because it shapes how reviewers think about the work.

Configure scoring

Give higher weights to the criteria that should have more influence on the final 0–100 quality score. Weights are shown as percentages and the scored criteria must total 100%.

Set the answer scores for each scored criterion:

For a number criterion, set the score for its minimum and maximum answer. Either end can be the high score—for example, 1 can score 100 while 5 scores 0. Answers between them follow the same direction.
For a yes/no criterion, assign separate scores to Yes and No. Either answer can score higher.
For a single-select criterion, assign each option a score from 0 to 100.
For an additive multi-select criterion, the selected option scores are added together, up to 100.
For a deduction multi-select criterion, scoring starts at 100 and subtracts the selected option scores, down to 0.

Use additive scoring when selected options represent positive outcomes. Use deductions when selected options represent defects or policy misses. An intentional empty multi-select answer scores 0 in additive mode and 100 in deduction mode; it is different from an unanswered criterion.

Choosing criteria

Common criteria include:

Accuracy
Clarity
Tone
Resolution quality
Customer effort
Policy adherence

Not every team needs every category. Pick the ones that reflect how your team defines good support.

Maintain scorecards over time

Scorecards should change when your operating model changes.

After a scorecard is used by an audit, its target, criteria, answer scores, and weights are locked so every result keeps the scoring standard it was reviewed against. You can still update its name or description, or archive it. Create a new scorecard when the scoring standard changes.

Update them when:

You launch new support channels.
A team adds new risk or compliance requirements.
Your escalation model changes.
Auditors keep disagreeing on the same criterion.

If a criterion causes confusion every week, it usually needs clearer wording or a better example.

How scorecards connect to QA review

Scorecards are the scoring standard behind audits. They help the team:

Review work more consistently
Compare quality over time
Spot weak patterns by topic, channel, or workflow
Decide what to improve in Knowledge, Flows, or routing

Without a scorecard, audits become much harder to compare across reviewers or across time.

Best practices

Start with fewer criteria than you think you need.
Use plain language in every criterion.
Train reviewers on examples before treating results as final.
Review the scorecard after the first few audit cycles.

Common mistakes

Too many criteria

This slows down reviewers and makes scores harder to trust.

Vague labels

If reviewers cannot explain what a criterion means in practice, it is not ready to use.

Mixing several ideas into one score

Keep each criterion focused so the team can tell what actually needs fixing.