# How do Zendesk for Customer Service teams tune AI answers after launch?

<p class="elv-tracking-normal elv-text-default elv-font-figtree elv-text-base elv-leading-base elv-font-normal" elv="true">AI that answers order status questions without an agent, pulling real order information rather than links to help articles, is what people value most in <a class="a a--md" elv="true" href="https://www.g2.com/products/zendesk-for-customer-service/reviews">Zendesk for Customer Service</a> for high-volume queues, so I've been looking at how teams tune those answers after launch. Going by G2 reviews, teams watch AI responses closely in the first weeks and adjust settings until answers run as intended, starting with repetitive questions where the right answer is clear. Finding where each setting lives takes a little getting used to on a large configuration, and that early attention is what keeps answers accurate for customers.</p><p class="elv-tracking-normal elv-text-default elv-font-figtree elv-text-base elv-leading-base elv-font-normal" elv="true">For <a class="a a--md" elv="true" href="https://www.g2.com/categories/help-desk">help desk</a> buyers, it comes down to how repetitive the ticket mix is. Queues with many similar questions, such as order tracking, see the biggest drop in agent workload. Teams with varied, complex requests still benefit, and spend longer checking AI responses before letting them answer on their own.</p><p class="elv-tracking-normal elv-text-default elv-font-figtree elv-text-base elv-leading-base elv-font-normal" elv="true">Ticket types worth comparing:</p><ol>
<li>Order status and tracking questions</li>
<li>Account and billing queries</li>
<li>Complex product questions</li>
</ol><p class="elv-tracking-normal elv-text-default elv-font-figtree elv-text-base elv-leading-base elv-font-normal" elv="true">Which ticket types did you hand to AI first, and how long did you check its answers before letting it respond alone?</p>

##### Post Metadata
- Posted at: 1 day ago
- Author title: SEO Content Specialist
- Net upvotes: 1


## Comments
### Comment 1

We started with order status questions and the drop in agent workload was noticeable. Checking answers closely early on was well worth it.

##### Comment Metadata
- Posted at: about 5 hours ago





## Related discussions
- [How well does Trello scale into a larger team?](https://www.g2.com/discussions/1-how-well-does-trello-scale-into-a-larger-team)
  - Posted at: over 13 years ago
  - Comments: 6
- [Can we please add a new section](https://www.g2.com/discussions/2-can-we-please-add-a-new-section)
  - Posted at: over 13 years ago
  - Comments: 0
- [Quantifiable benefits from implementing your CRM](https://www.g2.com/discussions/quantifiable-benefits-from-implementing-your-crm)
  - Posted at: over 13 years ago
  - Comments: 4