Fin tasks allow you to automate more complex processes with Fin. Often times, these processes may involve actions in your external systems that need to reliably follow your specific business rules (e.g. cancel an order, refund a subscription). Fin will trigger the task and be actively involved each step to resolve customer queries.

Here's what a Fin task might look like:

And the customer experience:

Check out Fin tasks in action:

Rather than using conventional workflows built by adding multiple steps—messages, branches, actions—you can set up Fin tasks using natural language.

You also no longer have to build complex decision trees to guide the user down the correct path based on their query; all of this can be handled by a simple description of a single Fin Task.

Using Fin Tasks, Fin can handle interactions that would otherwise require human action. For example, Fin can take a customer through extensive troubleshooting steps or analyze images of damaged orders from customers. These AI-driven enhancements reduce manual workload while ensuring customers receive fast, accurate, and intelligent responses—without needing human intervention

Fin tasks eliminate the need for customers to click through multiple buttons to reach the right solution. The entire experience feels much more natural and conversational, making it easier and more intuitive for customers to get the help they need.

For instance, you can instruct Fin to collect data as part of a task and instead of providing an input box for the user to type into, Fin will conversationally say something like “Can you please specify the subscription you are referring to?”

Both Data connectors and Tasks can be triggered by Fin, and they share a key similarity: You define a natural language description of when they should trigger, and Fin automatically activates them when a customer’s question matches. So, what sets them apart?

Data connectors: Simple, Single-Step API Calls

A Data connector is a single API call, best suited for straightforward, single-turn customer questions.

This requires a simple response—either the order is fulfilled or it isn’t. Fin can handle this by making a single API request to retrieve the order status.

Fin Tasks: Multi-Step, Logic-Driven Flows

A Fin Task (the focus of this article) is more powerful. It combines multiple actions (API calls) with your business logic, making it ideal for multi-turn customer questions.

This is more complex:

We recommend to use Data connectors for the simple use cases and Fin Tasks for the more complex ones.

Ask yourself: Which processes would you like to automate with Fin? Are there any existing complex workflows that could be streamlined using Fin tasks?

Focus on processes that will have the biggest impact on reducing support volume. A good approach is to identify 1 or 2 key processes to start with.

If you already have existing workflows that handle some of your support processes, you don’t need to create brand-new workflows to test this. Simply add the Fin Task block to your existing workflows to enhance their efficiency. As an example, your current troubleshooting workflow might look something like this:

While a Fin task block would simplify this and improve your customer experience dramatically:

Look for existing workflows where customers currently have a poor experience—especially those requiring multiple button clicks before getting an answer. These are ideal candidates for Fin tasks, making the experience more seamless and conversational.

Visit Train > Tasks and click on "New Task"

Every Fin task should be given a title telling Fin when to trigger the task. This is very important, so that Fin triggers it at the right time.

Similarly to the title, the description also tells Fin when it should trigger the task, in greater detail. This describes the situation Fin should use the task, provide example customer queries, and could also call out situations where Fin should not use this task.

Use this block to give Fin custom instructions for a task to perform. You can template the instructions with data, instruct Fin to run actions, store action response data and reply to users.

Actions can be configured to retrieve or update data in an external system. For more details, tips and best practices on creating actions themselves, go to this article. For more info on how to design and use your APIs with Actions, go to this article.

The response data from the action will become available as temporary attributes within the Fin Task (see next section).

Temporary attributes are useful when you need data only for the current task and don't want to save it permanently in Intercom for reporting or other purposes. You can create these attributes to hold information that's relevant to the specific task being executed.

For example you can create a temporary attribute to:

This new block pauses the task until it receives a webhook request, making it ideal for processes that require external input before moving forward.

Imagine your task needs to integrate with an external system—such as a third-party application for identity verification. With "Wait for Webhook," the task stops at this step until the external system sends a webhook response. Once received, the task resumes and takes action based on the provided data.

Examples

This feature ensures your tasks are seamless and synchronized with external systems, reducing manual intervention and keeping things efficient! To add a “Wait for Webhook” to your task, simply click “Add Step” and search for “Wait for Webhook”.

Inside the task, click on "Preview" and ask a question: