Quality Reviews with Auto QA
Auto QA helps QA teams review more conversations by scoring them against a QA scorecard (sections and questions).
Goal
Use Auto QA to: - find low-scoring conversations quickly, - understand which questions are failing most often, - coach teams using consistent criteria.
Step-by-step workflow
1) Start with the QA dashboard (if available)
- Set a time range (last 7–30 days).
- Identify:
- average QA score trend,
- distribution of scores,
- any spikes in failures.
2) Drill down to low-scoring conversations
- Click the low-score bucket (for example “QA < 80”).
- Review a sample set of conversations.
3) Review question-level details
In each conversation: - review the overall QA score, - review section/question outcomes (pass/fail/score), - read explanations for why a question was scored that way (if shown), - validate against transcript/thread.
4) Identify systemic issues
Look for recurring failures such as: - missing verification steps, - missing disclosures, - improper closing, - policy or compliance misses.
5) Coach and track improvements
- share examples with team leads,
- update coaching materials,
- watch trend changes over time.
Auto QA results view
Auto QA results are displayed on the QA tab in conversation details, showing:
- Overall QA score
- Section-by-section breakdown
- Question-level pass/fail results with explanations
Figure: Auto QA report showing scorecard results alongside the conversation transcript.
Providing feedback
If you disagree with an Auto QA result, you can provide feedback:
Figure: Feedback option to flag incorrect Auto QA results for review.

