| Announced on | Rollout starts | Rollout ends |
| May 14, 2026 | May 14, 2026 | May 15, 2026 |
We're introducing Reasoning for AutoQA in an early access program (EAP) to explain why AutoQA gives a specific grade or determines that a category doesn’t apply.
This announcement includes the following topics:
What is changing?
Reasoning is now available for AutoQA system categories, system spotlights, AI prompt-based categories, and AI prompt-based spotlights.
When you click an AutoQA-graded category score or spotlight in the conversation review or feedback panels, a Reasoning section expands and shows:
- The AI’s step-by-step evaluation logic.
- The specific evidence it identified in the conversation.
- For N/A ratings, why the category was determined to be not applicable.
Current limitations:
- Voice conversation evaluations aren’t yet eligible for this Reasoning EAP.
- Reasoning isn’t yet displayed when testing a prompt.
Why is Zendesk making this change?
AutoQA delivers automated quality evaluations across multiple areas, but you’ve told us you need more clarity on why specific scores or insights are given. Without this transparency, it’s harder to fully trust automated evaluations or take targeted action based on them.
Reasoning addresses this by turning AutoQA from a black box into an explainable system, making it easier to trust scores, calibrate expectations, and identify where manual review is warranted.
What do I need to do?
To learn about this feature, see Accessing and viewing conversations. If you'd like to be part of the EAP, turn on AutoQA Reasoning in Admin Center.
If you have feedback or questions related to this announcement, visit our community forum where we collect and manage customer product feedback. For general assistance with your Zendesk products, contact Zendesk Customer Support.