AI category detection is a beta feature available to selected customers only. If you're interested in joining the beta, please email Pavel at pavel.kozlov@intercom.io
Overview
AI category detection enables Fin to automatically categorize conversations based on their content. This helps streamline customer support by improving routing accuracy and reducing manual effort.
Key benefits:
Automated categorization: Fin uses predefined categories to automatically classify conversations, eliminating the need to run customers through a triage phase.
Seamless integration: AI category detection works across all Intercom channels and can be used in workflows, reporting, and inbox.
Improved efficiency: By using Fin to categorize conversations, teams can reduce response time and improve triaging accuracy.
AI Category Detection consists of three key components:
Extended conversation attribute of a “List” type - This attribute is used to store all categories and their descriptions. The "Description" enables Fin to identify the appropriate category based on conversation content.
“Auto classify attribute” workflow step - This triggers automatic categorization of a conversation at any point in the workflow, applying the relevant category attribute. This conversation data attribute can then be used in branches, eliminating the need for complex triaging processes and improving the accuracy of conversation routing.
AI Engine - Intercom’s patented AI Engine™ is the core component of this feature, responsible for selecting the most appropriate category based on the conversation’s content.
How to set up a conversation attribute for AI category detection
Before using the Auto Classify Attribute step in Workflows, you need to create a custom conversation attribute of the “List” type and provide the descriptions.
On the Conversations page, click Create attribute.
Click on the Format dropdown and select List.
Fill in the "Name and "Description" fields for the attribute, e.g.
AI Category - Examply features and products categorized for AI.
Fill in the "List options" section with clear names and descriptions. You can click Add option to add more options.
Click Save.
Writing effective AI categories
AI categories consist of two parts: a name and a description. The name should be a clear, short title, while the description provides detailed information about when to use that category.
These categories can serve many purposes, from sorting customer inquiries to detecting spam or analyzing customer sentiment.
Follow these guidelines to create effective AI categories
Create clear, concise names - Choose short, descriptive titles that immediately convey the category's purpose.
Write comprehensive descriptions - Include all relevant details about what belongs in the category. Think about every type of situation or content that should fall under this category and describe them in the description. Providing a detailed description will help Fin identify conversations that match this option. You can include keywords and examples of customer questions.
Make categories distinct - Avoid creating categories that overlap too much. Your categories should be clearly different from each other, making it easy to determine which one best fits a given situation.
Creating your categories - A step-by-step approach
First, list all the specific subcategories you want to include. Then, write a description that covers all these elements. This ensures your category descriptions are complete and practical.
Try passing these subcategories and their descriptions to a writing tool such as Claude AI or ChatGPT to consolidate them into the main AI category.
Here are some real-world examples
Sentiment Categories:
Positive - A positive sentiment means the user who wrote the message seems generally happy or satisfied and is probably feeling a positive emotion.
Negative - A negative sentiment means the user who wrote the message seems generally unhappy or dissatisfied and is probably feeling a negative emotion.
Neutral - A neutral sentiment means the user who wrote the message seems to be neither happy nor unhappy and it is difficult to guess their emotion.
Spam Detection Categories:
Spam - Automated spam that is sent to the customer support agents. This topic includes auto-responders, newsletters, guest posts, and other general spam messages that could be ignored by the CS analyst.
Legit - Legitimate conversations in which the user has an actual issue that should be handled by a customer support analyst.
Topic Categories:
Help Desk - The Help Desk topic covers a range of discussions related to managing systems, including automation rules, macro creation, email integration, team management, troubleshooting, and more.
Billing - Billing encompasses managing subscription plans, invoices, payment methods, discounts, plan features, trials, account restrictions, refunds, and more for a seamless billing experience.
Account Management - Account Management covers discussions related to user accounts, including account creation, deletion, updating personal and payment information, and more.
How to use AI categories in Workflows
After creating your conversation attribute, you can start using it in Workflows.
Go to Automation > Workflows.
Create a new workflow or edit an existing one.
Add the Auto classify attribute step anywhere in your workflow where you’d like your conversation to be categorized.
Select the AI conversation attribute you created.
In the workflow builder, you can use the conversation attribute you’ve created like any other Conversation Data Attribute. For example, you can add a branching condition based on the attribute list options and route the conversation to a specific team or teammate.
You can also use this conversation attribute in a Note or a Message step.
Save your workflow and then test your new AI-routing using the workflow Preview.
Set your workflow live.
To categorize a list attribute, each option in the list must have both a name and a description; otherwise, the attribute will not be selectable.
How to use AI categories in the inbox
Teammates working in the inbox can see when a list attribute has been set by Fin in three places:
Conversation events within the conversation thread.
Internal notes.
The inbox sidebar (when conversation attributes are pinned).
If Fin sets the wrong list option, teammates can correct it by selecting the appropriate option from the inbox sidebar. Please email us if this happens and we'll review it. 🧐
Optimize your AI categorization
Start simple: Begin with a few key attribute list options that cover the most common topics your team handles. You can expand and add more options as needed.
Refine your list attribute options: Regularly review the attribute options being selected by Fin to ensure they're accurate and aligned with your customers' questions.
Try the list attribute in reporting: Experiment with using list attributes in custom reports to gain insights into the types of questions your customers are asking.
We welcome your feedback on how useful this! It will help us improve the feature moving forward.
Need more help? Get support from our Community Forum
Find answers and get help from Intercom Support and Community Experts