Automatically detecting customer intent, sentiment, and language

What's my plan?

Add-on

Copilot

Intelligent triage uses artificial intelligence (AI) to enrich your Support and messaging tickets with actionable information without the help of agents or admins. Specifically, intelligent triage predicts the intent, customer sentiment, and language for new tickets. You can then use this information to automate workflows with triggers and automations, set up views for teams, create reports, and more.

This article contains the following topics:

• Requirements for using intelligent triage
• Understanding intelligent triage
• Turning on and configuring intelligent triage

Related articles:

• About intelligent triage
• Viewing and managing intelligent triage predictions

Requirements for using intelligent triage

Intelligent triage is available for organizations who meet the following requirements:

Requirement type	Requirement criteria
Industry This requirement applies only to intent predictions. You may still use intelligent triage for language and sentiment predictions without being in one of the listed industries.	Retail and e-commerce business-to-consumer (B2C). Software platforms or applications. Financial services. Traditional and online banking only. Insurance: Includes auto, health, home, and pet insurance. Employee experience: Includes IT and HR departments only. Travel, hospitality, and tourism: Includes airlines, bus lines, travel agencies, car rentals, and booking management. Entertainment and gaming. Includes online casinos, betting platforms, and ticket selling for events. Education. Includes information about education courses, examinations, admissions, scholarships, and student life.
Ticket volume This requirement applies only to intent predictions.	More than 1,000 tickets created by end users in the supported languages in the last 6 months
Ticket language distribution	At least 90% of tickets in a supported language
Right model fit This requirement applies only to intent predictions.	To determine how well the current machine learning model applies to an account, we perform a quality check in a sample of tickets from the past 6 months. During the quality check, intelligent triage must label at least 30% of tickets with high confidence and at least 60% with medium confidence.

Understanding intelligent triage

When you enable intelligent triage, the system adds new fields to your tickets:

• Intent: A prediction of what the ticket is about. To see the possible values, open the Intents page in Admin Center.
• 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, open the Language page in Admin Center.
• Language confidence: How likely it is that the language prediction is correct. Possible values are High, Medium, and Low.
• Sentiment: A prediction of how the customer feels about their request. Possible values are Very Positive, Positive, Neutral, Negative, and Very Negative.
• Sentiment confidence: How likely it is that the sentiment 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 update the field values 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, Language, and Sentiment fields, and are structured as follows:

• intent__value of Intent field
• language__value of Language field
• sentiment__value of Sentiment field

Tip: When creating triggers, automations, or reports, you can use the field value or the tag, depending on which one makes your build easier. For example, when creating a trigger with multiple intents, it’s easier to use tags because you can add them faster than adding separate conditions for multiple field values.

To learn how to create a report showing the results of intelligent triage, see Explore recipe: Intelligent triage predictions and confidence.

Tip: You can also use the Tickets API to return the intent, language, and sentiment as custom field values.

Understanding intent, language, and sentiment values

Intent

For the Intent field, intelligent triage is trained to identify intents. Additionally, intent and sentiment detection works only for the following languages:

Arabic Bulgarian Chinese (Simplified) Czech Danish Dutch English Finnish French German	Greek Hebrew Hindi Hungarian Indonesian Italian Japanese Korean Norwegian Polish	Portuguese Romanian Russian Spanish Swedish Thai Turkish Ukrainian Vietnamese

The Zendesk Intent Model includes intents for all of the supported industries. An account with this model has access to relevant intents and use cases based on their ticket conversation data. This model allows us to support accounts with multiple brands, mixed use cases, and industries that cross over between our currently supported industries.

Language

For the Language field, intelligent triage can detect approximately 150 different languages. See Zendesk language support by product.

Sentiment

The Sentiment field is a prediction of how positive or negative a customer feels about the request they’re submitting. The prediction is made based on the text of the first message in the ticket. If configured to do so, the sentiment is updated every time the end user replies.

Sentiment is grouped into one of the following categories:

• Very positive: The message likely contains strong positive words (like “brilliant” or “perfect”), positive words modified by intensity adverbs, or multiple positive sentences.
• Positive: The message likely contains phrases expressing gratitude, or one to two positive sentences.
• Neutral: The message likely contains factual statements without additional negative quantifiers (like “any” or “always”), or a mix of positive and negative statements.
• Negative: The message likely contains phrases expressing frustration, complaints containing negative words, or repetition of the same dissatisfaction.
• Very negative: The message likely contains strong negative words, capitalized text, multiple exclamation marks, or multiple negative phrases.

Intelligent triage is specifically calibrated for customer service. This means that a ticket isn’t assigned a negative sentiment just because a customer has an issue with their order, can’t find the information they need, or some other similar “negative” situation. Instead, the model is tuned to analyze sentiment with the assumption that the customer is contacting customer service because they have an issue that needs to be addressed.

Turning on and configuring intelligent triage

Admins can turn on intelligent triage in Admin Center. By default, this feature is turned on.

Intelligent triage adds predictions to tickets that were created only after it was turned on. For other reasons you might not see predictions on tickets, see Why didn't intelligent triage add predictions to a ticket?

To configure Entity detection, see Automatically detecting unique information in tickets.

To turn on and configure intelligent triage

1. Go to the page for the prediction type you want to configure:
   • In Admin Center, click AI in the sidebar, then select Intelligent triage > Intent.
   • In Admin Center, click AI in the sidebar, then select Intelligent triage > Sentiment.
   • In Admin Center, click AI in the sidebar, then select Intelligent triage > Language.

   The status of the Intent, Language, and Sentiment prediction types is on by default.

2. On the prediction type's page, deselect the Detect <prediction type> checkbox if you don't want the system to automatically enrich tickets with that prediction type.
3. On the prediction type's page, click Manage Settings.
4. Choose from the following options:
   • (Sentiment only) Dynamic detection: Select the checkbox to update the sentiment based on the latest end user message in the conversation.

     

     Note: Dynamic detection is supported for email channels only.
   • Channels: Select the email channels you want to turn on.

     The Web form, Email, and Web service (API) channels are turned on by default. You can also turn on the Text, Web Widget, Mobile SDK, Mobile, Facebook Post, X Corp, and Social Messaging channels. (If you create tickets via channel integrations, select Social Messaging.)

     Note: While we generally don’t recommend turning on additional channels, you might consider enabling them if you have high ticket volume on those channels. Channels with more conversational messages, system notifications, or spam will have lower-confidence predictions.
   • Messaging channels: Select the messaging channels you want to turn on.
     Note: If you turn on messaging channels for the Sentiment prediction type, note that dynamic detection is supported for email channels only.

     The Web Widget, WhatsApp, and Facebook Messenger channels are turned on by default. You can also turn on the Android SDK, Google RCS, Instagram DM, iOS SDK, LINE, Slack, Telegram, Twilio SMS, X Corp DM, Viber, WeChat, Apple Business Chat, Google Business Messages, KakaoTalk, MessageBird SMS, Native Messaging, and Sunshine Conversations API channels.

     Note: Turning on any one of the Web Widget, iOS SDK, Android SDK, or Native Messaging channels automatically turns on all four channels together.
   • Exclusion conditions: To exclude tickets created by agents from being enriched with the prediction type, select the Ignore agent-initiated tickets checkbox.
     Note: When this checkbox is selected, suspended tickets that are recovered are not enriched with intelligent triage predictions.

     

5. Click Save.

Note: If you use any third-party tools to make automated updates to your tickets, make sure those tools are configured to use the safe_update property so that intelligent triage ticket fields aren’t inadvertently overwritten. See Protecting against ticket update collisions.

Displaying prediction types in the ticket header

For the Intent and Sentiment prediction types, you can configure if the context or sentiment that's detected is displayed in the ticket header of the conversation.

See Viewing intelligent triage predictions to learn about the agent experience.

To configure prediction types in the ticket header

1. In Admin Center, click AI in the sidebar, then select Agent copilot > Auto assist.
2. Scroll down to the Ticket context section.
3. Select the Show intent in the header checkbox to display the context that's been detected in the conversation.
4. Select the Show sentiment in the header checkbox to display the sentiment detected in the conversation.

   If you've turned on Dynamic detection for Sentiment, the sentiment will update in the header based on the latest reply from the end user.

   

5. Click Save.