Automated triage uses artificial intelligence (AI) to enrich your tickets with actionable information without the help of agents or admins. Specifically, automated triage uses AI to predict the intent and language for new tickets.
Automated triage enables the Zendesk ticket fields for intent and language predictions and confidence, which you can then use for setting up views for teams, automating with triggers or automations, and enhancing Explore reports.
This article contains the following topics:
Understanding automated triage
Automated triage in Zendesk Support
When you enable automated triage, the system adds four new fields to your tickets:
- Intent: A prediction of what the ticket is about. To see the possible values, you can open the ticket field and check the Field values list.
- Intent confidence: How likely it is that the intent prediction is correct. Possible values are High, Medium, and Low.
- Language: A prediction of what language the ticket is written in. To see the possible values, you can open the ticket field and check the Field values list.
- Language confidence: How likely it is that the language prediction is correct. Possible values are High, Medium, and Low.
These fields are populated via ticket updates by Zendesk’s machine learning model based on the content of the first message of a ticket when it’s submitted. Agents can change the Intent or Language fields if necessary.
When the fields are populated, tags are also automatically added to the ticket to help you build triggers, automations, and reporting. These tags reflect the values in the Intent and Language fields, and are structured as follows:
- intent__value of Intent field
- language__value of Language field
To learn how to create a report showing the results of automated triage, see Explore recipe: Automated triage intent and language predictions and confidence.
Automated triage with the API
Additionally, enabling automated triage also enables the Ticket Predictions API. You can use this API to deliver the same type of intent and language predictions for a given ticket.
The API returns three values for intent and confidence, as opposed to the single value added to each ticket field.
When you use the API, the confidence for intent and language is presented as a score between 0.0 and 1.0. These scores correspond to the High, Medium, or Low values in the ticket fields:
- High: >=0.8
- Medium: >=0.5 and <0.8
- Low: <0.5
Enabling automated triage
Administrators can enable automated triage in Admin Center.
To enable automated triage
- In Admin Center, click Objects and rules in the sidebar, then select Business rules > Automated triage.
- Select Turn on automated triage.
- Click Save. The Channels section appears after a
few seconds, with Web form, Email, and Web service
(API) enabled by default.
- (Optional) If necessary, select any additional channels that
automated triage should populate intent and language values for. Note: While we don’t recommend enabling additional channels, if you have a lot of ticket volume on other channels, you might consider enabling some of the additional options below. Channels with more conversational messages, system notifications, or spam will have lower-confidence predictions.
- Twitter DM
- Twitter Like
- Get Satisfaction
- Web Widget
- Mobile SDK
- Help center post
- Forum topic
- Closed ticket
- Ticket Sharing
- Facebook Post
- Facebook Private Message
- Satisfaction Prediction
- (Optional) If you want tickets created by agents to be included in the
automated triage process, click Exclusion conditions and clear
the Ignore agent-initiated tickets checkbox. This checkbox is
selected by default.
When testing your automated triage settings, Zendesk recommends clearing this checkbox, or else using an end-user account for testing. If you don't, test tickets won't be automatically triaged.
safe_updateproperty so that automated triage ticket fields aren’t inadvertently overwritten. For more details, see Protecting against ticket update collisions.
Allowing agents to see and update automated triage fields
When you enable automated triage, the system automatically fills out the Intent and Language fields, but these fields are visible in tickets by default only if you have a single ticket form.
- If you have multiple ticket forms and you want agents to be able to see and change these fields in tickets, edit your ticket forms and add the new fields by dragging them into the form.
- If you have a single ticket form but you don’t want agents to be able to see and change these fields in tickets, edit your default ticket form and remove the new fields by dragging them off the form.
- If you want to use the automated triage field values only for Explore reporting or API use, you don’t need to add them to any ticket forms.
If you opt to let agents see the Intent and Language fields in tickets, they can update the values in these fields if they feel they’re not correct.
Remember that Zendesk’s machine learning model is intended to detect the language and intent of a ticket’s first message only. Agents should understand that any updates they make to these fields should still be based on the first message.
Intents and languages detectable by automated triage
For the Intent field, the machine learning model used for automated triage is currently trained to identify intents for e-commerce organizations specifically. Additionally, intent detection works only for the following languages:
For the Language field, automated triage can detect approximately 150 different languages.