Announcing latest enhancements to Zendesk Answer Bot

 We’re excited to announce several small but significant improvements we’ve been able to bring to Answer Bot. With these enhancements, Answer Bot will be able to deliver better suggestions and make it easier for admins to set up business rules using ticket tags. 

The two enhancements we released are 

• Automatic Answer Bot ticket tags for trigger setups
• Extending 1-word search behavior to all Answer Bot channels 

Automatic Answer Bot ticket tags for trigger setups 

This enhancement introduces 4 new ticket tags that are automatically added to Answer Bot tickets submitted via email and webform.  Tags are added based on the end user action and the suggestion outcome. The below table can be used to understand when these tags are added.

Ticket tag name	When is it added?
ab_suggest_false	Added when Answer Bot is successfully triggered but failed to find any matching articles
ab_suggest_true	Added when Answer Bot successfully made an article suggestion
ab_resolved	Added when the end user indicated that the suggestion was helpful
ab_marked_unhelpful	Added when the end user marks the suggestion as unhelpful

With these tags, support admins are now able to more easily create custom ticket views and business rules such as:

• Creating a view of tickets where Answer Bot was unable to suggest an article
• Creating a view of tickets where Answer Bot resolved the tickets
• Creating a view of tickets where any ticket where Answer Bot has made a suggestion are excluded
• Automatically following up when customers have self-solved
• Following up when answer bot wasn’t able to find a matching article

If you’ve been following our best practices, you may already have your Answer Bot trigger(s) configured to add a tag such as “answer_bot_fired” to any ticket that Answer Bot attempts to assist with.  One issue with this approach is that this tag is being added to every ticket where the triggers fires, which also includes tickets where Answer Bot was unable to find and recommend an article.  With this enhancement, your tickets will be automatically labelled with more granular identification of whether or not Answer Bot successfully suggested an article.  Additionally, you can see if the end user has provided feedback indicated if the article allowed them to resolve their issue.

To learn more about how to set up Answer Bot support triggers, please refer to Best practices: Setting up Answer Bot triggers, views and workflows.

Extending 1-word search behavior to all Answer Bot channels 

Overall, we’ve found that Answer Bot’s AI-powered recommendations are more accurate and relevant than a keyword search, especially when the question is asked as a full sentence. 

However, there are times when a good old-fashioned keyword search works better. For example, when a user asks a single-word question via Web Widget, Answer Bot defaults to using a keyword search, as this is generally more accurate for single-word queries. 

As a part of the release, we have extended 1-word Help Center search behavior to both Answer Bot API and Answer SDK allowing more channels to enjoy the benefit of this solution. 

Check out this article for more on how Answer Bot works with your content.