Overview

AutoQA allows you to speed up your evaluation process by reviewing and scoring several or all of your scorecard's criteria and scale the number of evaluations you do.

There are two ways to use AutoQA when evaluating tickets, Co-pilot and Autonomous.

Co-pilot

The "Co-pilot" option can be used both in workloads and stand-alone evaluations. In this case, as soon as a ticket is selected for evaluation, the AI will review and score the ticket and the evaluator can review and adjust AI's result before submitting the evaluation.

With the "Co-pilot" option, the AI will only score the criteria that have AutoQA enabled. This helps reduce the time it takes for your evaluators to review tickets.

Autonomous

If you don't want to have evaluators reviewing tickets manually, you can select the "Autonomous" option.

This option is only used in workloads. As soon as a workload cycle is created for AutoQA, it will start reviewing tickets randomly based on the workload configuration.

In "Autonomous" workloads, the AI will score all the criteria in the scorecard, whether AutoQA is enabled or not!

Dashboard

To see the tickets that have been evaluated with AutoQA, go to AutoQA > Overview.

Evaluations

This graph shows you the number of tickets AutoQA has evaluated over time.

Original and corrected average score

This graph shows how the average ticket evaluation scores change before and after correction.

• The yellow line represents the original average score.

• The blue line represents the corrected average score.

• When the lines are close together, it means AutoQA evaluations are well aligned with manual evaluation.

• When the lines are far apart, it means manual corrections are significantly adjusting scores.

  • Learn how to optimize AutoQA here.

Mean absolute error and corrected evaluations

This graph shows how aligned AutoQA scores are with manual review scores.

• The yellow line represents the mean absolute error, the difference between the AI’s initial evaluation and the corrected (human-reviewed) score.

  • Lower percentages mean the AI’s evaluations are more accurate.

  • Here, the difference between the AutoQA score and the corrected score is 0.05%.

• The blue line shows the percentage of corrected evaluations.

  • Here, 0.42% of all AutoQA tickets were corrected on September 28, 2025.