Feature Request for Instant AI agent answers: Establish Mapping between ticket form and Help Center Category | Community
Skip to main content
Idea Submitted

Feature Request for Instant AI agent answers: Establish Mapping between ticket form and Help Center Category

Related products:Zendesk AI & automation
  • February 26, 2026
  • 6 replies
  • 0 views

Josef11

Please give a quick overview of your product feature request or feedback and note who in your org is affected by this issue

We are referring to the announcement “Announcing instant AI agent answers in web forms

In our Help Center, we provide articles for multiple product families. While these products operate within the same general domain, they differ significantly in terms of functionality, configuration, and support requirements.

Each product family is represented by its own dedicated Help Center category. Customers submit their inquiries via web forms, and each product family has its own specific ticket form.

For us to effectively leverage Instant AI agent answers, it is essential to establish a strict mapping between the ticket form and the corresponding Help Center category.
 

In other words:

If a customer submits a request for Product Family A using Ticket Form A, the Instant AI Agent must retrieve answers exclusively from Help Center Category A. Including articles from other product family categories would result in irrelevant or misleading suggestions, ultimately reducing answer quality and customer trust.

 

This requirement primarily affects our customers (who rely on accurate instant answers), but also impacts our support agents and administrators. If AI suggestions are not precisely scoped, agents will face increased clarification effort, and customers may lose confidence in the feature.

 

A category-based or form-based content restriction mechanism for Instant AI agent answers would therefore be critical for us to adopt this feature at scale.
 

What problem do you see this solving?

The goal is to enable customers to receive fast, targeted support through Instant AI agent answers, thereby increasing both our self-service rate and overall customer satisfaction.

For this to work effectively, it is essential that the AI-generated answers are derived exclusively from articles specifically created for the relevant product family. If the system pulls content from the entire Help Center without restriction, customers may receive inaccurate or misleading information due to differences between product families.

Precise content scoping is therefore critical to ensure relevance, maintain trust in the AI responses, and truly improve resolution speed and self-service adoption.

 

When was the last time you were affected by this lack of functionality, or specific tool? What happened? How often does this problem occur and how does this impact your business?

We are currently affected on an ongoing basis, as we are not able to fully leverage the feature in its current implementation.

 

Due to the lack of precise scoping between ticket forms and corresponding Help Center categories, we cannot activate Instant AI agent answers in our web forms. If we did, customers submitting a request for Product Family A could potentially receive suggested answers from Product Families B, C, or others. From a customer perspective, this would be confusing and would undermine trust in the quality and relevance of the provided solutions.

 

This limitation occurs every time we evaluate enabling the feature - which effectively means it impacts us continuously. As a result, we are unable to realize the intended benefits of higher self-service rates, faster resolutions, and improved customer satisfaction.

 

From a business perspective, this means:

  • Lost opportunity to deflect tickets through effective self-service
  • Continued higher workload for support agents
  • Reduced ability to scale support efficiently across multiple product families

Until precise content scoping is possible, we cannot responsibly roll out this functionality to our customers.

 

Are you currently using a workaround to solve this problem?

At this time, we are not aware of any viable workaround.

Due to the missing capability to restrict Instant AI agent answers to specific Help Center categories based on the selected ticket form, we cannot safely enable the feature. As a result, the only current option is to refrain from using it altogether.

 

This means the limitation cannot be mitigated operationally or through configuration in its current state.

 

What would be your ideal solution to this problem? How would it work or function?

Our ideal solution would be the ability to explicitly link a ticket form to a specific Help Center category within the Instant AI agent answers configuration.

 

Functionally, this would work as follows:

  • Each ticket form (e.g., Product Family A) can be mapped to one or more defined Help Center categories.
  • When a customer submits a request using that ticket form, Instant AI agent answers would only generate responses based on articles within the linked category.
  • Articles outside of that scoped category would be excluded from consideration.

This form-to-category mapping would ensure that AI-generated answers are contextually accurate and product-specific, preventing irrelevant cross-product suggestions. It would allow us to confidently activate the feature while maintaining answer quality, customer trust, and operational consistency across multiple product families.

6 replies

Josef11
  • Author
  • February 26, 2026

One additional enhancement to the ideal solution would be the ability to not only link a ticket form to specific Help Center categories, but also to selected sources within Zendesk Federated Search.
 

In an optimal setup, the configuration would allow us to define a precise content scope per ticket form - including both designated Help Center categories and explicitly selected Federated Search sources. This would ensure that Instant AI agent answers draw exclusively from the most relevant and product-specific knowledge repositories, further improving answer accuracy and consistency.


Marcin15
  • February 27, 2026
Thank you for taking the time to provide us with your feedback. This has been logged for our PM team to review. For others who may be interested in this feature request, please add your support by upvoting this post and/or adding your use case to the comments below. Thank you again!
 

  • March 2, 2026

We have the same issue, however we use a slightly different approach at this moment. We have one ticket Form and on the ticket form the user has to select a product (Custom Object). In our Helpcenter we also have a seperate category for every product with all articles related to that product. 
We want te prevent that AI is using articles from the wrong category (product) to give answers back to the customers. 

How can we dynamically redirect (configure) AI to use the correct helpcenter content for a Ticket (form) / Chat / …


Josef11
  • Author
  • March 3, 2026

@erik18 
Thank you for sharing your detailed scenario - we can very much relate to this.

In our setup, we also use Zendesk ticket forms primarily to differentiate at the highest level between product families. Within each ticket form, we typically refine the categorization further by product and, depending on the selection, by component and/or module.

However, to enable truly precise and context-aware AI-generated answers, we would likewise require an additional level of granular mapping between ticket form fields and the corresponding Help Center categories and/or subcategories.

A more fine-grained association would significantly improve the relevance and accuracy of instant AI agent responses, especially in complex, multi-product environments.


Alexander40

This feels like one of those cases where relevance matters more than raw article count.

 

When AI answers search across too broad a knowledge pool, even a technically related article can still be wrong in context.

 

A fallback pattern that usually makes sense is:

- use the ticket form or product selector as the first scope

- search the matching product family/category first

- only widen the search if confidence is low

- make the source scope visible so admins can debug bad suggestions

 

Otherwise it’s hard to tell whether the problem is article quality, taxonomy, or search scope.

 

Curious whether people here want strict scoping by default, or search narrow first and then expand?


Alexander40

This feels like one of those cases where relevance matters more than raw article count.

 

When AI answers search across too broad a knowledge pool, even a technically related article can still be wrong in context.

 

A fallback pattern that usually makes sense is:

- use the ticket form or product selector as the first scope

- search the matching product family/category first

- only widen the search if confidence is low

- make the source scope visible so admins can debug bad suggestions

 

Otherwise it’s hard to tell whether the problem is article quality, taxonomy, or search scope.

 

Curious whether people here want strict scoping by default, or search narrow first and then expand?