Fin Product Area

Hi! I am curious to learn more about how Fin ingests long form content that is not necessarily written in a way to support it (i.e Content chunking, formatted for Ai readiness). Looking to learn if Fin has the ability to ingest something like an owners manual and able to answer any questions related to it without creating a specific support article on a topic related to the owners manual. 

Can we better contextualise Help Centre content for Fin?For example, having internal-only guidance/instructions embedded directly within articles that end users cannot see. While content/source guidance and snippets help, they become difficult to scale when managing thousands of articles and hundreds of supporting snippets.In our case, Fin’s errors are often less about incorrect content and more about misunderstanding nuance or scenario context. Being able to add contextual instructions, edge cases, or operational guidance directly alongside the article content would make maintenance significantly easier and likely improve response quality.

Hi!I am oftenly using Intercom while browsing other Chrome tabs.When I am focused on one issue, and focused on another tab opened to find a solution to it, I find it very distracting to have the intercom tab name switching non-stop to tell me about notification from other conversations. The name of the tab moves every 10 seconds from the name of the open conversation to notification about other messages coming in. Is there a way to block it? I do not have preferences on what appears as the name of the tab, for as long as it stays still. NB : setting my status as away does not work. Thanks a lot for your help.   

Hey Community,we have issues setting up Intercom AI so it solves different Topic automatically. Can you help us?For some reason FIN suddenly and randomly answers with wrong answers.We tried everything (with and without guidance,...) All the Content stayed the same and in the past FIN was super helpful. We are looking for an expert to help us automate different Topics in our Customer Support.Thank you in advance!

We’d like the ability automatically assign a team inbox when a teammate initiates an outbound conversation, ideally based on the From email address used.For example, a conversation sent from billing@company.com should automatically belong to the Billing inbox from the moment it is sent, regardless of whether the customer replies.Currently, teammates must remember to select an inbox manually. If they miss that step, outbound conversations can sit without a team inbox, creating gaps in visibility, ownership, and follow-up.Requested behavior: Allow admins to map each outbound From address to a default team inbox. Apply the assignment when the initial message is sent, without waiting for a customer reply. Preserve an inbox explicitly selected by the teammate. Provide a fallback team inbox for outbound conversations without a matching address rule. The core requirement is that every teammate-initiated outbound conversation has an assigned team inbox immediately, so none are left witho

Hello!Is there a way to create a “subject line” for macros? Seems only the email body can be created so I wanted to see if anyone knew if that was possible. Thank you so much, appreciate the help. 

Our company offers software which contains databases our end users need for their work. The databases sometimes use difficult terminology that is hard to interpret for our users. In an attempt to improve the ‘readability’ of the databases, I have recently uploaded a bunch of ‘metadata’ associated with the most used databases in our software - breaking FIN as a result.Prior to starting the upload December 2025, I had contact multiple times with my Intercom account manager to determine the ‘optimal’ way to upload a large amount of qualitative (text) data to Intercom. The data on our end originally is stored in JSON format - a format we can’t directly upload directly to Intercom. After some discussions, we opted to split the JSON file into smaller .txt files, which we manually uploaded as ‘snippets’ to Intercom. We have spent a considerable amount of time setting this up. Now the system is ‘stuck’ since Christmas 2025, with no viable solution yet in place. No new content we upload is inge

Hello,Based on the documentation (https://www.intercom.com/help/en/articles/3894-using-intercom-with-content-security-policy), it appears that both unsafe-inline and unsafe-eval need to be added to our Content Security Policy in order for the JavaScript widget to function.From a security perspective, this raises serious concerns, as it increases the attack surface. Our reporting endpoint shows that unsafe-inline isn’t actually required by your code, but unsafe-eval is being triggered in the vendors-app-modern.js file at the following line:var mod = eval("quire".replace(/^/, "re"))(moduleName);Is there any way to eliminate this eval call? If not, we may need to reconsider our use of Intercom due to these security implications.Best regards,Salim

We have a number of products with a lot of overlapping terminology, and a lot of audiences that overlap with other audiences. This has made it difficult to target the information Fin uses very specifically. We have been considering setting audiences up using conversation data instead of just user data, but I’m having a hard time understanding if/how that would work. We have a new conversation attribute that would be selected and help define the audience for that conversation. But is that effective? If the customer asked another question in the same conversation, or had chosen something incorrectly, could the audience for that conversation ever be changed? Open to any suggestions here on how this could work, or what better options might be. 

we would like to forward all incoming e-mails sent to our support team to intercom. this can be achieved by using the default inbound address from the e-mail forwarding settings in intercom. no problem. BUT: we do have 2 teams in our support:one team for all clients in france (french speaking only) 1 team in germany (german speaking only) we are using 2 e-mail addresses for our clients:france-support@company.comgermany-support@company.com if i forward all imcoming e-mails to my germany-support@company.com e-mail address to the intercom default inbound address, everythin is OK - i receive these e-mails as new chat in the intercom inbox. BUT: we are usinig 2 inboxes in intercom:GERMAN support teamFRENCH support team can i create 2 default inbox addresses, so that this would be possible: client in France sends an e-mail to france-support@company.com --> will start a new chat in the France Inbox in intercomclient in Germany sends an e-mail to germany-support@company.com --> will sta

The search bar on the modern Help Center template only runs once a query has at least 3 characters. Intercom confirmed this limit is fixed, not configurable, and the same for every language.A 3-character minimum works for English, where 1–2 letter inputs rarely carry meaning. In Korean, Japanese, and Chinese, a single character carries much more meaning, and many of the most common support search terms are exactly 2 characters:- Korean: 플랜 (plan), 가격 (price), 결제 (payment), 환불 (refund), 계정 (account)- Japanese: 料金 (fee), 解約 (cancellation), 請求 (billing)- Chinese: 价格 (price), 退款 (refund), 账户 (account)Today these queries return nothing. Customers assume the article doesn't exist, then contact support or give up. Adding a particle (e.g., 플랜이) makes search work, but customers have no way of knowing this.Requested change (any of these would solve it):1. Lower the minimum to 1–2 characters for CJK queries2. Apply the minimum per Help Center language3. Make the minimum length configurable in Hel

The search bar on the modern Help Center template doesn't return results for queries under 3 characters. For example, searching "플랜" (plan) or "가격" (price) returns no results, while "플랜이" or "가격은" (the same root word with a particle attached) works fine.This minimum length may work reasonably for English, where 2-letter inputs are often low-signal, but it's overly restrictive for Korean, where 2-character words are frequently complete, high-value search terms. A language-aware minimum, or a lower threshold for CJK queries, would better serve non-English Help Centers.Could you let us know if this minimum length is configurable, or if there are plans to make it language-aware?

Hello,  We are starting out with intercom and struggling to verify outbound messaging, I used auto configuration and everything got updated in Cloudflare, but I am still seeing Warning on Outbound Messaging, here is what I am seeing https://ibb.co/s94ZrnSq (Not sure why I cannot upload image here)I can manually configure it in cloudflare, but I don’t know what are TXT record values, is there anyway to see those values so I can manually change in cloudflare?ThanksKets  

For a long time, keeping Fin up to date on bugs our customers hit was slow and manual. Bug tracking has to be quick and accurate, and it has to scale, and our old process wasn’t any of those. Then I learned about Fin Attributes. Here’s what we built (which costs us $0 extra to run) needs very little maintenance and keeps humans fully in control, plus the walls I hit along the way.The problemWhen a customer reported a bug we already knew about but Fin didn’t, Fin went in circles. It repeated the same troubleshooting steps and effectively told the customer the problem was on their end. Keeping Fin informed meant hand-updating Snippets for every bug, which took ages and didn’t scale.The ideaCould we sync the bugs we’re tracking straight into Fin, so Fin knows about them, escalates them, and even gets those conversations linked to the right tracker ticket so we can update customers when they’re fixed?Our first version used the API to pull every Bug tracker ticket into a single Confluence p

Hey Everyone!We have data of conversation, conversation parts, contact, and company to our datawarehouse, extracted using Intercom API.Now, we are trying to replicate one table that our client has built in the Intercom Reports which basically shows the conversation with bug tag across the companies. It’s a really basic table, containing the following columns:Conversation ID Company ID (from Company Standard Attribute) Company Name (from Company Standard Attribute) Conversation Tag Some more additional fieldsNow, we are able to create the most of the fields from the API data. However, we are facing the issue for the Company ID and Company Name.For example, in Intercom report table, a single conversation has 8 companies (as comma separated string) in the Company ID and Company Name. While in datawarehouse we have around ~52 companies.For the datawarehouse logic, we are using contact as the a bridge to map the conversation data with company. In datawarehouse the flow is Conversation → Con

When an agent starts a new outbound email (Inbox → + → Conversation → Email), the ID field in the sidebar shows “—” until the email is sent. Our agents need the reference number while writing the email, to log it in our back office. Today they have to send first and then go back to copy it.The only way to get an ID before sending is to create a Customer ticket first, which is much slower than a simple email.Request: give a new outbound email its conversation ID as soon as it’s started, and show it in the sidebar before sending, the way Freshdesk shows a ticket number from the start.

Why is there no option in workflow builder to Expand all paths, similar to Minimize all paths option?There is only Expand all paths in view, but we have to scroll horizontally and click on it several times to expand all the steps and have a complete overview.

Hi everyone, I'm running into an issue while configuring the Intercom Messenger widget to trigger based on dynamic session states. We recently added the standard JavaScript snippet across our website to offer quick help for users encountering setup or download issues, but the chat bubble frequently fails to load for visitors using specialized Chromium-based desktop wrappers.Alongside the initialization hiccup, user attributes like operating system and client build numbers are either delayed or drop out completely before reaching the conversation inbox. Has anyone experienced similar script conflicts or timing issues when running Intercom on a dynamic web application, and is there a recommended way to delay the boot script or adjust Content Security Policies to ensure stable tracking?

Hey everyone 👋 I'm an indie founder building ChatShield, a tool that helps support teams identify potentially risky conversations across their workspace not just AI chatbot replies, but human-agent conversations too. For example, an agent might accidentally promise an unauthorized refund, make an incorrect pricing commitment, or share information they shouldn't. I'm trying to understand how Intercom teams currently catch these issues. A few questions: - Do you manually review conversations, use internal QA, or have another process?- Are mistakes by human agents as much of a concern as AI-generated replies?- Would automated flagging of conversations for human review be useful to your team? I'd genuinely appreciate hearing how you're handling this today. I'm also looking for a few Intercom teams willing to test ChatShield and give honest feedback. Product: https://chatshield.cloud/ Thanks!

TextSpeed WhatsApp Business API helps businesses connect with customers through automated, secure and scalable WhatsApp communication. Use the API for notifications, customer support, promotional campaigns, order updates, authentication and chatbot conversations while creating personalized customer experiences at scale.

 Hi everyone 👋 🎅 I wanted to share this with the community because I couldn't find a single article on how to get Fin ready for Christmas. There's plenty out there on getting support teams ready for the holidays, but nothing on the AI agent itself.I think this is really important, so I'm sharing it now while there's still time before Black Friday. It's the playbook I use myself, it's all practical, and every step shows where it lives in Intercom. Fingers crossed it makes it onto the Intercom blog one day 🙂 Your team is ready for Christmas, but is your AI agent? Every year we see the same "how to prepare your support for Christmas" guides. Hire temps, write macros, extend your hours. It's good advice, but it was written for human teams, and in December Fin is the first one most of your customers talk to.So for me the real question is a different one. Does Fin know your Christmas cut-off dates, does it tell people honestly when a human is back, and does it feel even a little bit like

Hey! I discovered that the Side conversations option allows to send messages to Slack—neat! However, it is not straight forward to later on find the conversations that have side conversations. Any ideas of how this could be done? In short, this is the workflow that we are aiming at: When we need to ping someone from the team for a quick question, start side conversation and set message to selected Slack channel (this part is covered) Then find a way to easily keep track of this conversation (or any conversation that as a side conversation) within Intercom, like using Views or something similar. (This is not done!)I figured it could be done using Views filters (like when we filter conversations that have tickets, etc.), but I failed to get it working. Any thoughts, tips or tricks?

It would be wonderful to have a Intercom native feature where you could set up a alert threshhold feature on a chart that informs dedicated stakeholders if the volume was hit (without requiring periodic review). Examples might be: - If CX score hits X% - If conversation volume exceeds X conversations - Etc. We have a lot of busy stakeholders managing multiple different tasks, and things can fall below the radar. But this way you could draw peoples attention, without them having to cut out time daily to review dashboards to check performance across different releases.

I have a question about WhatsApp.How do you deal with the fact that we can't send active support after 24 hours of conversations with no customer interaction?We stopped, for example, on Friday, with that, with that, we were unable to respond to conversations on Monday    

Sharing how we're using Slack side conversations, plus one problem I'm hoping someone can confirm or work around.What we've builtOur technical support team has a private internal Slack channel with an always-on AI agent sitting in it (built on Claude, running in our own agent framework). There's no trigger word or slash command — you just start a new thread in the channel and the agent replies in that thread. It has read access to our internal systems (Jira/Confluence, GitHub, Sentry, our own product data), so answers come back grounded in real context rather than generic. It's become the fastest way for the team to get a sanity check mid-ticket.From Intercom, we can open a Slack side conversation on a ticket and post a question straight into that channel. That half works perfectly — the question lands, and the agent answers in the thread.The gapOnly human replies flow back into the side conversation in Intercom. The agent's replies never appear. I'm assuming that's because they're pos