Fin Product Area

I know the default survey uses emojis with hover-over text labels, but it seems like our customers might not be understanding the difference between each rating (see attached image). It’s starting to impact our metrics a bit.I’d really like to keep using Intercom’s native CSAT tool for reporting and widgets, but I’m wondering: Is there any way to customize the default emoji options (e.g., swap them out or show the text labels instead)? Or do we know if there’s already a product request or feature idea out there for that? Also related question, is it possible to tweak how CSAT is calculated in our instance? For example, could we have “OK” responses count as positive? A lot of our users are parents or older family members, and based on some past examples, “OK” seems to represent a satisfied response rather than a neutral or negative one.

When I checked the Japanese inquiries in Fin AI's Performance, I found that all the content is displayed in English even though customers wrote in Japanese. Can you change it to as it was written?Also, could you add a feature to filter and select only Japanese customers?When I search in Japanese in the content from Knowledge, the relevant items do not appear in the results. As my customers use only Japanese, I would really appreciate if you could solve this issue. Thank you.

I am currently setting up the Fin AI Agent. I have a question regarding the "Talk to a human" flow.When a user triggers the handoff to a human agent, how are the notifications handled for the assignee (or the team inbox)? Does the assignee receive an immediate email notification? Is it purely an in-app notification/browser push? I want to ensure that when the AI hands the conversation over, our support team is alerted immediately, even if they aren't staring at the Intercom dashboard.Thanks in advance!

Today we noticed an issue with Fin failing to respond to tickets. There have been no changes to the workflow but for whatever reason it just hasn’t responded to queries. We noticed the increase in FRT and drop in involvement but looking back we are almost certain this has happened before. It appears to be intermittent and no real obvious answer but curious if anyone else has experienced this and how you resolved. 

​@Nathan Sudds Are the office hours from yesterday recorded with the Q and A I can share with my coworker? TIA, Melissa Speed

Using the existing Intercom integration with Jira, I find it to be lacking in functionality.For example, a support agent reports on an issue and the agent raises a customer ticket or a tracker ticket. This action triggers an automation process that creates a bug in Jira.The important link between any support platform and Jira, is the AffectsVersion and fixVersion.I need the support agent to report in Intercom on the Affected Version, which should map to the Affects Version field in Jira.In return, once the development team assigns a fix version to the bug, I require the fixVersion field in Jira to map to a Fix Version field in the Customer or Tracker ticket.Such field doesn’t exist, so I have to create it… but I see no way to map the fixVersion Jira field to any field in Intercom tickets.Appreciate any insights on this matter - or how any of you communicate the Affects Version and Fix Version between your development work to support tickets? 

Hi Community, I’m successfully sending WhatsApp template through Series, but no conversation is being created in Outbound. Can you assist if I’m doing anything wrong?

We would need to add Content Security Policy to our application. I saw this article on Intercom Help. There it is stated to use 'unsafe-inline' but I am wondering is there a way not to use it, since we would like avoid that in our policy?

Hi, im trying to send a message when a client does not answer to a bot. I´m sending a few options in buttons when the conversation starts, and I would like for the bot to identify if the customer hasn´t responded in 30 minutes and send them a “Are you still there?” message, then wait other 30 minutes and close the conversation if there still hasn´t been answered. Is something like this possible? Fin tells me you can´t send automatic messages if the client hasn´t responded to a conversation with the bot.

We’re hoping to gain insight into how other customers are forecasting handle time, particularly when using Intercom data. We've noticed discrepancies in the handle time metrics reported by Intercom that don’t align with the structure of our operations.Typically, I export the raw data and filter out outliers manually. However, we continue to see a significant number of conversations with unusually long handle times that lack a clear explanation. This inconsistency makes it difficult to trust the accuracy of the reports.We’re looking for a better approach—either a clearer understanding of how Intercom calculates handle time behind the scenes, or alternative methods for forecasting handle time more reliably using Intercom data.Thank you in advance for any guidance or shared experiences!

We are new integrating Intercom with our systems. Right now having problems with location object... Where can i get some list of valid (countries, regions, cities) in order to map correcctly?

Issues:Once you add a content to a missing one and create either a snippet or article from it- you can never go back or see an audit trail of what you have actually done.    There is no list anywhere where you can see which content gaps you have actually optimized. We would like to store them somewhere which them might be helpful for configuting another area   Once you add the content and hit save, it opens the next suggestion instead of taking you back to the Optimize page, this is quite confusing and unrelated from what you were doing.    No export buttons anywhere   There is a table where they give percentages of conversations, but you can not click open any of these to see where they are stemming from- again, also no export option located.   ​​​​​​​ 

I want to embed an interactive demo into our Help Articles from another platform. Can I do that? The code is: <div style="position: relative; padding-bottom: 56.25%; width: 100%; height: 0;">    <iframe      src="https://tashi.storylane.io/demo/swdjjjrzseopquot;      allowfullscreen      style="position: absolute; top: 0; left: 0; width: 100%; height: 100%;"    >    </iframe>   </div>  

Hi everyone,I’ve been trying to resolve an issue where Fin AI continues to touch or count spam emails as unresolved conversations, even though I’ve created a custom spam workflow that should handle them before Fin ever activates. We also have the Intercom Spam Filter turned on and was advised to not disable it (it only catches upto 4 spam tickets a day, it doesn’t work so well with our spam emails). Here’s what’s happening: I built an “Aggressive Spam Filter” workflow that uses branches with “message contains” and “email domain” rules to automatically close spam emails. Fin is not supposed to be involved, but it still logs actions like language detection or “junk/spam” decisions after my workflow triggers. Fin initially sets the language, my workflow closes the email, Fin reopens the email and reports it as spam. Because Fin touched it, it’s logged as unresolved.  These conversations then appear as unresolved Fin tickets, which is hurting our Fin resolution rate (some days up to 80

My company is anxiously awaiting the ability to access Fin through a channel API in our own bespoke messenger. Are any of you already doing something similar? If so, how’s that going?

Hi, I’m not sure if this has been addressed before, I couldn’t find an existing topic. Is there a way to manually create a user or lead in Intercom without entering an email address? I understand we can use either an Email or User ID for new users, but in this case, our agent needs to start a WhatsApp conversation with a lead who doesn’t have an account with us yet. The lead was referred through another channel and has no prior conversation history, so there’s no record in our Intercom database. Since the customer doesn’t have an email, we’re looking to manually create a lead using only their phone number.!--endfragment>!--startfragment>

Hi Community! We are still newish to Intercom, launching in earnest in April of 2025, and have been slow-rolling it out to our users. We currently have Fin enabled for ~100 of our customers, with plans to double, then triple that number in the upcoming weeks.Overall, Intercom has been great and Fin is performing even better than our initial expectations. That said, the one reoccurring issue we continue to see is users shortcuttong/circumventing Fin--when the chat starts, their first question is “Talk to a person” or “create a case,” and since Fin is frictionless, he will not fight them and defers to the case handover process (we use ServiceCloud). Issue is, since no question or context is present, none of the AI-detection fields fill in, and the case just is sent to a ‘triage’ queue until our reception team can reach out to determine what the users’ question/issue was so they can then place them into the correct Support queue (we do not currently offer live-chat support). We’ve bounced

Hi,I would like to implement a workflow which is triggered when a conversation/ticket is created in a specific inbox, that triggers an api configured in data connector to fetch the information. Can this information sent by the api be added as a note to the ticket/conversation as shown below?How is the dataconnector’s success response data be sent as a note?  

Hello there!I’m looking for a to conduct quality assurance on the back-office ticket submission process and track the responsible parties for each ticket. Is there a way to generate a report that captures that info? Currently, we have a boolean attribute that we can toggle, however, we can’t filter through it by the person who submitted it. Any help is greatly appreciated! 

Fin is connected to my work Instagram business page. A follower recently tagged me in a story (which I received as a DM), and Fin is answering that DM. How do I stop that from happening?

Hi there,We’ve implemented Fin AI (Chat and Email) on our support desk, and it’s been fantastic so far—it’s doing a great job handling first-line queries! 😊However, we’ve noticed a challenge when a customer submits a vague message. Fin will respond with technically correct information, but it’s often not relevant to what the customer truly needs.I’m wondering if it’s possible to improve this by getting Fin to ask clarifying questions before responding. I’ve been advised to set up a new workflow where specific vague questions prompt Fin to ask follow-ups, but trying to predict and account for all the potential vague queries customers might come up with feels inefficient.Is there a better way to manage this? Any advice or insights you could share would be hugely appreciated!Thanks so much

Hi!We are not using Fin anywhere in our product and are not currently interested in implementing it. It’s never been integrated into a workflow and I have never manually turned it on. From our customer’s help chat perspective, they are seeing the following header on the chatbox: It happens when the agent they are speaking to goes idle or if the user clicks on a conversation that was previously closed. Why does this show up? I need to remove this. Our product’s customers have no idea what “Fin” is. I also don’t know what “The team can also help” means.Any help would be so appreciated! Thanks

Hello,I want to add Intercom remote MCP to an AI agent. How do I add my callback URI to the allow list for the remote MCP?Thanking in anticipation.

GET https://js.intercomcdn.com/vendor-modern.1ffdfa4a.js net::ERR_SSL_PROTOCOL_ERROR GET https://js.intercomcdn.com/frame-modern.f3182eef.js net::ERR_SSL_PROTOCOL_ERRORPlease let us know if you need more details

Hey,We’ve got following message when trying to push the e-mail to our users:"Email can't be sent right now as your app is being reviewed. We've emailed your app admin with more details."What is approx time to have this review completed?I really need to have the solution working properly, and it does not. Best,SentiOne Team