All Collections
Fin AI
Fin AI Agent
Getting started
Set up and go live with Fin AI Agent
Set up and go live with Fin AI Agent

How to set up and go live with Fin AI Agent using Workflows for control and flexibility.

Beth-Ann Sher avatar
Written by Beth-Ann Sher
Updated this week

This article covers how to configure and set Fin AI Agent live using Workflows, providing you with control and flexibility over how Fin behaves.

You'll need to have access to Workflows on your plan and be using Fin AI Agent (on trial or paid). If you don't have access to Workflows on your current plan, you can still learn how to customize and set Fin AI Agent live here.

Customers using Profiles:

We'll soon be deprecating Profiles and Workflows will become the main way to set up and use Fin AI Agent, so we encourage you to remove or migrate any Profiles to Workflows if you haven't done so already.

Get started

Go to Fin AI Agent > Fin Overview and open the Set up and go live tab.

Now you can either choose:

Make sure you've finished adding your content to Fin AI Agent first.

Option 1: Quick setup

This allows you to quickly set Fin AI Agent live by simply configuring two options.

First, decide who will see Fin. You can allow Fin AI Agent to provide answers to your users, leads, and/or visitors.

Click + Add audience rule and use data attributes in Intercom to target your audience further. For example, Fin could be enabled for conversations with users on your free plan.

Then specify what action should be taken If Fin cannot resolve the conversation. You can choose to pass to team and assign to a particular team or teammate.

Or, choose to close conversation and offer alternative ways for the customer to get help. For example, provide a contact number or link to another resource.

Once you're done, click Review and set live to see what this Workflow looks like.

Select the Preview button at the top of the builder to experience an interactive conversation with Fin AI Agent using your Workflow.

When you're ready, click Set live 🎉

If you'd like to tailor this workflow further, click Edit and you'll be taken to the Workflows builder where you can use the steps below to customize the experience.

Option 2: Tailor your own workflows

You can also add Fin AI Agent to an existing Workflow, or start with a template.

Start with a template

Choose Start with a template to maximize Fin AI Agent's visibility while still providing a robust handover experience to your teammates.

Once you've finished setting up your trigger rules and scheduling, you can start customizing Fin AI Agent using the steps below.

Add Fin AI Agent to an existing Workflow

Selecting Add Fin to an existing workflow will take you to your Workflows where you can open and edit one of your prebuilt Workflows to include Fin AI Agent.

The best Workflow triggers to use with Fin AI Agent are:

  • Customer opens a new conversation in the Messenger - use this if you want Fin AI Agent to immediately introduce itself in the Messenger and respond to customers' questions.

  • Customer sends their first message - use this if you want Fin AI Agent to respond to customers' questions across alternative channels (WhatsApp, Instagram, SMS, etc.)

Add Fin AI Agent to your Workflow by creating a path and selecting Let Fin answer.

This inserts a new Fin AI Agent step which can then be configured to control the behavior and handover.

Configure Fin AI Agent behavior

Choose which channels Fin AI Agent is enabled on

To enable Fin AI Agent on particular channels or for a specific audience, click on the Workflow trigger settings (first block in the Workflow builder).

Channel selection is only available when using triggers such as "Customer sends their first message" or "Customer sends any message".

To start customizing the Fin AI Agent experience for your customers and teammates, click on Let Fin answer from the Workflow path.

Set expectation for human support

If you don't want Fin AI Agent to hand over to a teammate, you can configure the step to not set any expectations for human support.

This will change the Talk to a person 👤 button to Get more help 🌐 and will prevent the conversation being assigned after inactivity.

Auto-close abandoned Fin AI Agent conversations

Choose when to auto-close conversations (if at all). By selecting the checkboxes, you are telling Fin AI Agent to auto-close conversations in these scenarios if the customer is inactive for the designated time.

You can also specify exactly how long Fin should wait before a customer is considered inactive and the conversation is closed.

Customize what Fin AI Agent says when closing the conversation

When a customer goes inactive after interacting with Fin, you can make sure they got what they needed by sending a chaser message which lets them know they can still speak to your support team, even if the conversation is being closed.

Let Fin collect additional context

Gather more information with Fin AI Agent when a customer asks to speak to the team.

By prompting the end-user for more context before Fin hands over to an agent, there is another opportunity to get more and/or better information than presented the first time around. The benefit to this is two-fold:

  1. Gives Fin a chance to answer: many times users don’t give the AI Agent a chance, asking to speak to an agent before even asking Fin a question

  2. If context is provided, even if Fin can’t provide an answer, the agent will spend less time gathering context and will be able to troubleshoot faster

Note: Insider knowledge 🔍 Before launch we tested this feature extensively—observing an increase in answer rate, confirmed resolutions, and CSAT.

​Teammates will see a section Ask for more information before handover in their Fin Workflow so they can toggle it on or off 👇

Ask for a conversation rating when Fin AI Agent finishes

At the end of a customer's interaction with Fin AI Agent you can ask them to give a conversation rating by toggling this setting on.

Control content sources

If you've set up Custom Answers, you can choose whether Fin AI Agent uses Custom Answers or AI Answers or both.

Fin AI Agent always gives precedence to Custom Answers before using AI Answers in a conversation. Learn more.

When only Use Custom Answers is selected, the option to Only suggest one answer becomes available.

Configure Fin AI Agent handover

Fin AI Agent will attempt to answer questions posed by customers. If the customer clicks That helped 👍 , Fin AI Agent will close the conversation and mark it as resolved.

This doesn't prevent the customer asking follow up questions if they want.

If the customer clicks Talk to a person 👤 they will follow a handover path that you will need to configure in the Workflow.

For example, you could set expectations for when your team will reply or collect additional information that will help route the conversation to the right team. This helps make the transition as smooth as possible and can save teammates' time and effort when picking up the conversation.

Learn how to use the Workflows builder.

Pass to a reusable Workflow

You can also pass to a reusable Workflow if you have a common triage or data collection process that you want it to follow. Use the Pass to a reusable workflow step to hand Fin AI Agent over to another Workflow.

Audience targeting

By using branching, you can create different Fin AI Agent experiences for different audiences.

For example, for paying customers, you might want to hand over to a teammate when Fin AI Agent can't answer but for non-paying customers, you might want to direct them to your Help Center.

Set live

Now that you've finished setting Fin AI Agent up in your Workflow, you are ready to preview and set the Workflow live. This can be done in the top right of the Workflow builder.

You'll need billing permissions to set it live if the T&Cs for Fin have not yet been accepted yet. Accepting these terms is mandatory to be able to use Fin AI Agent.

Once Fin AI Agent is live, you’ll be able to view:

Attachment icon


Need more help? Get support from our Community Forum
Find answers and get help from Intercom Support and Community Experts

Did this answer your question?