Fin Product Area

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

Hello all, we have a number of use-cases where we need to engage with 3rd parties for “shared customers” in a form of “ticket sharing”.Current practice is to use a side-conversation and trigger an email (which is likely a conversation part) to that 3rd party, and then responses from that 3rd party link back to the side conversation.Whilst the above sort of works, it is not the most intuitive experience as it is not obvious that there has been a response (you have to open each ticket, navigate to side conversation, see a traffic light indicating a new email).The bigger challenge is this does not consistently work (it does with Zendesk as it was levelling feature to match what they do), but does not for SPRINKLR and Service Cloud where the responses to side conversations create a whole new disconnected conversation, resulting in additional handling and bringing together of conversations.Has anyone done much work on how to connect 3rd party systems and “share tickets”? Is side conversatio

Test Environment:OS: Windows 11 version 25H2 (OS Build 26220.7523)Browser: Edge (Version 151.0.4129.15)Screen Reader: NVDA (2025.3.2), JAWS (2026.2512.50), NarratorURL: https://help.enlyft.com/en/articles/11242176-enlyft-web-browser-extension-download User Impact:Users who rely on screen readers cannot understand the purpose of the iframe because its name is not announced. This makes the video content inaccessible and prevents users from deciding whether to interact with it. Pre-Requisite:Turn on Screen Readers: NVDA: Ctrl + Alt + N JAWS: Ctrl + Alt + J Narrator: Ctrl + win + Enter Verbosity: NVDA: Default JAWS: Beginner, Highest Narrator: Default Repro Steps:Open the URL: https://help.enlyft.com/en/articles/11242176-enlyft-web-browser-extension-download in edge browser in windows. Press down arrow key to navigate to the YouTube video in browse mode. Verify whether Screen reader announces the name for iframe present on YouTube video in windows or not.Actual Result:Screen reader do

Hi community 👋Has anyone found an effective way to handle customers who open a conversation by simply typing “agent”, without giving Fin / the AI Agent a chance to help?I’ve tried:general guidance encouraging self-serve first (minimal impact), and enabling “Set expectations for human support”, which unlocks “Ask for information before handover.” However, in our case the bot does ask a follow-up question, but then hands off to a human agent before waiting for the customer’s reply.What I’m looking for is a smarter approach (likely via guidance/behavior rules) where, when someone just types “agent” up front, the bot:probes for key details first, attempts to answer appropriately, and only escalates if it can’t be resolved.Any best practices, example setups, or wording that’s worked well for you would be hugely appreciated 🙏🏼

Enable users to view a public article in the Help Center while logged in, and via a direct link without being logged in, without having to duplicate the article→ Users : can view the article through the Help Center if they are connected→ Can also view the article through a direct link our team shares before they sign in or if they are not logged in 

I want to be able to check if making a change to content will improve Fin’s answer to a question without first having to publish it. Is there anyway to do this? 

I’m trying to setup an onboarding checklist in which multiple steps will launch product tours. I have the first step setup and functioning fine, but the subsequent steps/product tours I cannot seem to configure to launch when/where it is needed. Here’s the issue:Our customer’s URLs are : subdomain.domain.com …..When a customer creates their account and logs in for the first time, they are brought to their homepage. The checklist appears and the first item in the checklist launches a product tour on that page. There’s no dynamic aspect to that URL, so we can simply setup the “Use your tour everywhere” setting to use {subdomain}.domain.comThat tour has them create a case/project in their dashboard, which generates a new page/URL. The issue arises here, because that URL is different every time, for every user, however, the URL itself is not unique to a user, meaning any user in that dashboard (which there are typically multiple) can access that case/project/URL. The URL ends up being some

We recently integrated the Intercom Messenger widget on our support portal at cepcutapk.com to assist visitors with troubleshooting guides and template downloads. However, several mobile visitors browsing through our tutorials for capcut video projects are reporting that the chat launcher freezes completely or takes over ten seconds to render when loaded within Android in-app webviews.In addition to the UI lag, automated bot workflows fail to trigger during these sessions, and custom attributes passed via the JavaScript boot script seem to drop unexpectedly when users switch browser tabs. Has anyone experienced similar messenger initialization delays on content-heavy mobile sites, and is there a recommended deferral or initialization workaround to maintain performance?

 Hi everyone,I am opening this feature request to address a recent and highly crucial update from Meta’s WhatsApp Business Platform ecosystem: the official support for WhatsApp Coexistence.During my interactions with the Intercom support team, there seems to be a misconception that using the WhatsApp Business App and the Cloud API simultaneously on the same phone number is either unofficial or a workaround. I want to clarify that this is now a fully official feature documented and supported by Meta, and having Intercom support it natively would be a massive game-changer for many businesses.What is WhatsApp Coexistence?Meta now officially allows businesses to register and use the same phone number concurrently across the WhatsApp Business App (mobile/web) and the WhatsApp Business Platform (Cloud API).When configured correctly via Cloud API, this setup enables: The Best of Both Worlds: Centralized multi-agent support inside Intercom via the API, while maintaining the ability to make/rec

Hello everyone 👋I'm curious to learn what redaction tools you're using for masking PII (Personally Identifiable Information). At the moment, we're using Strac to redact all PII in real-time, but I'm interested to hear if anyone else uses different redaction tools integrated with Intercom.Would love to hear your experiences and any recommendations you might have!Thanks in advance! 😊

What I want is that when we write an article and we need to make sure a specific section is always highlighted/referenced by AI, we can indicate thisEx: certain actions are only possible if the client is working with a specific partner

Hi everyone,I'm hoping someone has run into this before.We're currently migrating our internal Knowledge Base into Intercom for Fin/CoPilot, but we're not manually creating each article in Intercom. Instead, we're uploading our existing documentation from Google Docs (exported as .docx) or as PDFs. We don't use Notion, Guru, or Confluence.The problem is that every upload loses a lot of its formatting, making the articles much harder for agents—and Fin/CoPilot—to read.I've already tried pretty much everything I can think of: Using Normal text only (no Heading 1, Heading 2, etc.) Using bold text instead of Word heading styles Keeping section labels (Purpose, Process, Before You Begin, etc.) as bold text on their own line Using bold for step titles (e.g. Step 1: Verify the order) with normal text underneath Keeping formatting very simple with basic bullet points Uploading both .docx and PDF versions The formatting still gets stripped or flattened during import.Has anyone success

When a caller has a hidden/private number, Intercom Phone cannot receive the caller ID and therefore has no phone number available for a callback.Today, these callers can still be presented with the option to request a callback, even though the callback cannot be completed. This creates a confusing customer experience and may lead customers to believe they will receive a callback that is technically impossible.I would like to suggest adding a condition/check before the queue-or-callback choice is presented. If no caller ID is available (anonymous/private number), the callback option should automatically be hidden and the caller should only be offered the option to remain in the queue.This would help avoid failed callback requests, reduce customer confusion, and create a more intuitive phone experience.

I’m working on a PC gaming website where visitors often have questions about games, installation, system requirements, and troubleshooting.For those using Intercom, what’s the best way to organize these types of questions? Would you recommend using a Help Center with separate categories for game information, technical issues, and general FAQs, combined with automated replies for common questions?I’d also be interested in hearing how other websites use Intercom to reduce repetitive support requests while still providing a good experience for users.

For teams doing this reconciliation today — is it entirely internal, or does anyone use an independent third party to verify a sample of Fin-reported resolutions against backend evidence?I’m especially curious where the Fin outcome has a commercial consequence: refunds, cancellations, account changes, or outcome-based billing.And when the backend evidence disagrees with the reported resolution, does that feed into what gets treated as a successful/billable outcome?

Running Zapier from google drive to intercom to add information into user notes, however it will keep piling on instead of editing the existing notes, and it becomes a huge eyesore and can be very confusing to new staff. I have played around but can't find any way to remove a User Note. Any way around this?

Is there anyone I can talk to about the new style or new structure of Intercom’s/Fin’s public articles? I noticed the left side bar and other things. Can we also do that with our public articles?

We’re going to stop using Fin AI. It’s a shame because the product itself is really good, but the pricing isn’t suited to marketplace business models like ours.With margins of around €1 per user per month (incl. VAT) and high volumes, the cost quickly becomes unsustainable.If possible, it would be great to share this feedback with your product teams internally — there’s likely a broader pricing vs. business model fit issue for marketplaces.

We've all spent time improving Fin's knowledge, testing answers and looking at resolution rates.But I'm curious about what happens after Fin actually sends the answer.Say Fin tells a customer:“Yes, you can get a full refund.”…but your actual policy doesn't allow it.Or it gives an answer that sounds completely normal, but contains something your team shouldn't be promising.How are you catching those cases today?Manual QA? Monitors? Sampling conversations? Customer complaints? Or do they mostly get missed?I'm particularly interested in teams running Fin at meaningful volume. How are you handling this in practice?

Hello folks! I’m trying to workout how to (if possible) create a follow-up Ticket when a NPS Survey is below a certain threshold.  The NPS Survey is currently being sent up to once every 90 days, and my intention is potentially follow-up with user’s based on their response.  I want agents to be able to review the NPS Survey response, and optionally contact the user based on the feedback given in the response.  Therefore, I was thinking the best way to go about this would be to create a ticket containing the NPS Score and NPS comment, and then the agent will be prompted to review (and optionally contact) the user.  However, I can’t work out how to get this to work via a Series; I can trigger a webhook (but that doesn’t seem capable of creating a ticket). I feel like I could probably use an Action, but I can’t work out how to trigger an action via Series.  Has anyone come across anything simila

Hello! Since view columns are stored locally in the browser’s cache, and clearing the cache resets the view columns' data to default, I developed a Tampermonkey userscript for my team to preserve selected/ordered table columns per view. Yesterday and today, it seems tests (or changes) have been conducted that caused the userscript to stop working, and the columns keep getting reset to default.Yesterday morning, the localStorage key was split into multiple different keys, causing our userscript to stop working. After a few hours into our day shift, the original localStorage key came back, and the userscript was working again. Today, the same symptoms were witnessedIs it expected that these changes made it to our environments? I’m more than happy to share more information if needed.

We have unique URLs for every page for a user. For example, User A home page URL is: toddleapp.com/platform/32476/unitPlansUser B home page URL is: toddleapp.com/platform/99172/unitPlans Will product tour work here?

ive added the intercom chat widgte to my website, but it isnt appearing on the page. the website loads normally and ive checked that the intercom code in included. when i open the website link, the chat icon doent appear, is there a setting or url-related issue i should check?

Fin doesn't read your knowledge base the way a person does. It reads structure, tags, and outcomes. Get those wrong during migration, and resolution rate suffers from day one — even if every article technically "moved."Across Intercom's customer base, Fin averages a 67% resolution rate. Top-performing teams hit 80% or higher. The gap between those two numbers usually comes down to migration quality, not the model.How Fin actually reads your knowledge baseFin doesn't treat your help center as a flat pile of articles. It uses hierarchy — Collection, Section, Article — to understand scope and context before it answers.Most help desks only support a three-level structure. Intercom allows a simplified Collection > Article path with no Section in between. That's fine natively. It's a problem in migration: if your source platform's categories don't map cleanly onto that structure, a migration tool has to guess, usually by dropping orphaned articles into a default folder. That default bucke

self‑only performance view in Workforce  (So an agent will only see their performance an not the whole team) - I believe this is in progress as per conversations with the Workforce team but just looking for an update.