Tips from Expert Users

New Support Stack episode: I sat down with ​@Brian Branca , AI Conversation Designer at Fin, on when a Fin Procedure is actually the right tool (vs when your knowledge base should have handled it instead).A customer told Brian's team they'd hit a really high resolution rate, then added "by the way, we haven't added any procedures yet."We also got into Operator now it's generally available. It'll take you from a blank canvas to a first draft in a couple of minutes (it built eight conditionals and the sub-procedures for me while I did the thinking). What it can't know is what your customers actually expect from you, so that still all yours.Watch Brian's first episode now

Hi all! This is my first post in this community 👋🏻Since we started with Intercom, we have had problems with customers saying that they have emailed us several times, but not receiving a reply. Some times, we have found out that they’ve misspelled the email or such. But recently, we have had several reports where they have emailed the correct address, but it hasn’t reached our inbox at Intercom. We have asked our tech team for support, but they can’t see no error on our side. The Fin AI-bot is just telling us to check our settings in our email domain. Has anyone else had this problem, and how did you fix it? This problem has made us start thinking about choosing another vendor at this point 🙈

New Support Stack Solo episode: teams get their Fin AI agent to around 50% resolution and then plateau. Adding more help articles doesn't move it, because you've run out of questions content alone can answer.So I walked through the three stages of how a customer can interact with an AI agent — informational, personalised, action-taking — what "stage one done properly" actually requires before you're allowed to move on, and when to reach for an attribute versus a data connector.If you haven't got engineering time for data connectors, you can still get Fin taking actions by putting a "loop in a teammate" step inside a Fin Procedure. Fin gathers the information, your teammate does the one click, Fin closes the loop. I show a real subscription downgrade going through that way. >> Watch it now << 

Someone at Fin got a worrying Slack message: a customer had used one of their internal feature flag names in a conversation. Had Fin leaked it?​@Dawn, AI Knowledge Manager at Fin, answered that in under ten minutes with Operator — searching every article, snippet, data connector, procedure step and guidance rule at once, then every conversation from that company. Turned out the customer had heard it from their CSM on a call.New Support Stack episode, and it's a good look at using Operator as a detective rather than a writer. I also throw in the trick I use to turn a conversation that went wrong into a simulation, so my procedures tell me when I've broken them (oooooops). Watch it now!  

New Support Stack episode: I sat down with ​@Dawn, AI Knowledge Manager at Fin, to talk about the Operator feature most beta users miss — memory.She walks through what to save (style guide, content readiness framework, naming conventions), what to keep out (vague rules, product info that already lives in your knowledge base), and the "echo principle" for FAQs, shown live on a real article. If you own Fin content or a help centre, it's a properly practical watch!Watch episode 15 here 

Hello everyone! 🇬🇧  I’ve got some cracking news to share with you all today. I’m officially launching a brand-new series of video tutorials here in the Intercommunity: Fin: Industry. The goal is simple: instead of talking about AI in the abstract, I’m going to show you exactly how to build and deploy Fin for specific, high-pressure sectors. This is a continuous series, so keep your eyes peeled for more coming your way! Episode 1: The Gambling Industry & Live Sports Data ⚽️🏇 In this first video, we’re diving into the fast-paced world of gambling. I explain how you can take Fin to the next level by connecting it to live sports data. Why is this a game-changer? In gambling, information is everything. If a punter is asking about live odds, match status, or "Why hasn't my bet settled yet?", they don't want to wait for a human. They need the facts now. By hooking Fin up to real-time data feeds: Instant Accuracy: Fin can pull the exact score or horse racing result the second it happens

Hi Intercom team 👋,I’m currently setting up our Help Center and testing different layouts for the collections.At the moment, in the Grid layout, the collection icon is always displayed above the collection title. I’d like to achieve a layout where the icon is positioned to the left of the title, similar to a horizontal “card” or list-style layout, while still keeping a clean, visual overview of collections.I know it’s possible because I saw it on a help center made with Intercom. I’ve tested:Grid, List, and Minimal layouts Different icon formats and sizes (PNG, SVG, high resolution) Various Help Center customization settingsHowever, I couldn’t find any option to:change the icon position (top vs left) control the structure of collection cards beyond the default layoutsMy questions are:Is the icon position for Help Center collections fixed by design? Is there any way (native or planned) to display the icon to the left of the collection title? Are there any recommended alternatives or be

I’ve had great success over the last three weeks feeding Fin Conversation Transcripts that have a negative CX score, to our other internal AI chat tools, and working to improve Fin AI Agent immensely. (Our Fin Ai Agent Resolution rate has increase by 6% up to 54%+ over the past four weeks. 💪🏼The setup:Build a custom report that pulls Fin AI Agent conversations where the CX score was 1, 2 or 3. Pull transcripts for the conversations each week Train our internal AI Agent on all of the info it needs to guide me best What tools I have to train this Fin AI Agent (All different Guidance, categories Communication settings, etc) Trained on all of our products using both the Help Center that Fin uses, and our full internal knowledge base with even deeper product info and knowledge.  The Prompting:Next, I iterated on a good prompt that works smooth, and scales. I started out with a simple one conversation prompt.This is a support conversation that was auto-categorized as low CX score for the

One thing we have been looking to do for a while is give fin access to simple status lookups for policies where payment is pending. The issue we are currently facing is that payment takes 3-5 business days to post on our platform (no clue why) however i have access to a running PowerBI report from accounting that contains policy numbers and also flags when a payment has been submitted almost immediately. the flags operate as a simple yes/no and would greatly decrease the amount of inquiries we get about payment status. essentially all of our workflows require that the retailer give us the policy number up front so matching wouldnt be a big problem. unfortunately our development team is spread thin so im really just trying to hack together a solution as proof of concept that we can get fin access to this data. ive done some work with data connectors successfully but this isnt in my wheelhouse. im wondering if theres any sort of integration/apps in intercom that could make my life easier

If you watched the Fin Procedures launch today, you may be interested in Support Stack Solo – Episode 1. I just released it today, to help you answer the question: are you building the wrong Fin Procedure first?It’s a short look at:What makes a good Procedure When not to use one Why there should always be a clear path to a human How I’m designing a lead qualification procedure in my own workspaceIf you’re working with Fin Procedures right now, this should help you avoid unnecessary complexity.Watch it on YouTube: Support Stack Solo E01

If you've been using Intercom’s Fin AI agent for a while, you'll remember when Tasks were the way to control how Fin behaved. Procedures have changed the game — but not everyone's had the chance to get their head around what's actually different and why it matters.In this live session, we're cutting through the noise and looking at exactly what's new with Fin Procedures: how they work, how they compare to the old Tasks setup, and what this means for your AI support strategy going forward.This one's for support leaders and CX teams who are actively using Fin with Tasks, and want to get ahead by planning for Procedures.You can join via LinkedIn or via YouTube.

 Your product experience is like a net - but without the right touchpoints, customers can slip through gaps.  Empathetic touchpoints help keep them engaged.Imagine your product as a net, designed to hold onto customers throughout their journey. But like any net, gaps can form—moments where users feel overlooked, frustrated, or disconnected. These experience gaps can lead to churn, missed opportunities, and broken trust. In my last article,  ‘Why Speed Alone Isn’t Enough’, I shared how we used Compassionate Macros to scale support without losing the human touch. But customer conversations aren’t the only place where empathy matters.Every part of your product—from onboarding to system messages—shapes how users feel. The challenge? Automation at scale often strips away warmth, leaving interactions feeling transactional or impersonal.So how do you ensure your product doesn’t just function well but also feels human?It comes down to embedding empathy into every automated touchpoint—from micr

Hi everyone! Just sharing you a short success story when I helped a business owner integrate Fin AI to her Facebook Online Shop. Plus, I added some tips in the later part."Totus Tuus Shop" is a small business that was struggling to manage a high volume of repetitive product and order inquiries on Facebook Messenger. The owner needed a way to automate sales without losing the personal touch of her brand.By applying the Fin QA Flywheel→ Analyze, Train, Test, and Deploy—I built a system that transformed her inbox. Seeing the workflow handle a full customer journey, from "What's the price?" to "Here is my shipping address," gave the owner the confidence and consideration to fully implement Intercom in her shop. **Please see the attachment below for the recording of the “Preview” of the actual execution of my workflow**Disclaimer: Note that the preview is different “visually” compared to the actual Facebook messenger/page execution. That’s why I made a few tweaks in order to adjust with Fin

Hey Intercom-Community.I would like to turn off Fin answers for group conversations, where our customers customer, our customer and our support are in a group conversation. I know that Fin will provide an inital answer based on the incoming E-Mail. After that inital answer, I’d like Fin to stop responsing to any more incoming Mails. Has anybody set up a workflow/conversation attribute/something that might solve this issue?Thanks for your input!Best, P

Hi everyone!Hope this isn’t too off-topic. 🙂I recently discovered that it’s possible to connect Intercom directly to ChatGPT (at least with a Business license), and I’m really curious to hear how others, especially smaller support teams like ours, are using it to save time and work more efficiently.I haven’t had much time to dive deep yet, but I’ve started experimenting. So far, I’ve used the connector to scan recent conversations and surface common feedback patterns or recurring issues. I’ve also used it to gather feedback on specific topics. Another helpful use case has been content updates: when we launch a new feature, I can ask ChatGPT to review our help articles and point out where information needs to be updated.I’d love to hear how others are using the Intercom → ChatGPT connection. Any workflows, prompts, or clever tricks you’d recommend?

Discover  how to Use Fin Attributes for Escalations in Workflows & Tasks from Intercom Community Top Expert ​@Nathan Sudds 💡  

📊  When Your Data Doesn’t Tell the TruthHow Fin’s AI Categorization Turned Messy Data into Clarity Every CX or ops leader knows the feeling. You pull up a report, and at first glance the numbers look fine. But something feels off. The story in the charts doesn’t line up with the reality on the floor. The problem isn’t the report—it’s the tags feeding it. When tagging is inconsistent, the data turns unreliable. And unreliable data makes reports lie. That was the bind this team found itself in. They needed to spot product issues early—like which models were most likely to crack, which parts kept showing up damaged or missing, or which shippers were the least reliable. With that kind of clarity, they could protect revenue, get ahead of problems, and spare customers unnecessary pain. But with inconsistent tags, they weren’t seeing the full picture. The pivot: Stop Wrestling with Tags. Start Tracking Reality 🔁 Here’s the move: instead of drafting yet another tagging framework with sprawli

Hey everyone — curious to learn from other teams here!We’re building out a Case Closure Data Model as part of our Intercom + Fin AI rollout, with the goal of turning every resolved ticket into AI-ready data for automation, prevention, and knowledge training.Our current thinking includes adding structured closure fields like: Technology / Sub-Technology Problem Code / Resolution Action / Root Cause Repeatability / Escalation / Customer Effort KB Linked The intent is to improve data quality, reduce “Other” categories, and help Fin identify repeatable automation candidates over time.I’d love to hear how others are approaching case hygiene at closure: Which closure fields have been most valuable for your AI or reporting goals? How do you balance structured data collection vs. agent friction? Have you seen any creative ways to tie closure data to Fin training or KB improvement? Any best practices, examples, or lessons learned would be hugely appreciated.— Lance 

Cross-post from other group, but…...Loads of new features, some AI capabilities complimenting/replacing legacy features and many terminology updates. The Intercom glossary is being updated daily, but after some recent Intercom partner training, felt it worthwhile to post some of our notes on the Intercom Lexicon for those that are new to the platform or about to explore Fin. Check it out here: https://www.journeymapper.net/blog/intercom-glossary-101  

Imagine being the sole support agent for thousands of customers on a growing platform, personally handling every request. That was my partner’s reality at a startup years ago. The sheer volume was overwhelming, but every interaction mattered to him, and he poured genuine care into crafting thoughtful, personalized responses for every single user. Customers felt valued, but it came at a cost: long delays that led to frustration despite his best efforts. The pressure to speed up was constant, and leadership understandably wanted faster responses. This is the dilemma many companies face: speed vs. care. Most companies, when faced with this situation, choose efficiency. They prioritize speed, churning through tickets as quickly as possible—but at what cost? Responses often feel generic, irrelevant, and devoid of empathy, leaving customers feeling processed rather than valued. But my partner didn’t want that. For him, the choice felt impossible: prioritize speed and risk losing the quality

Hey there 👋 Milan here✌ For those of you who might not know me from the previous version of the community, a bit about myself:I’m the owner of the Sovaolo.com MarTech agency and a long-time Intercom expert. I’ve been working with Intercom for about 5 years (initially as a Product Marketing Manager in one of Intercom’s clients) and since I started working as a freelancer, Intercom was core of my offering. I have delivered on more than 100 Intercom projects! Enough bragging, I’d like to share some of my tips on how to use Intercom like a Pro. 💪 Tip #1: Preparation is the key to successful project deliveryMy own drawing of “Touchpoints between Intercom and Visitor/Lead/User during their journey”Explaining what Intercom can do, to all relevant stakeholders, explain it in whole even if relevant stakeholder is not present (as the words might reach them). I have my own drawing of “touchpoints between Intercom and Visitor/Lead/User during their journey” (image above). I have that with the ex

If you really think about it, it makes total sense:You want the audience of any message to as specific as you can in order to provide the best fit between audience and message. You are probably already filtering your messages by language and other attributes from your product.In today's landscape, you need to convince future customers that you can deliver what you promise. One way that works for many companies is to use social proof from existing customers (on review sites or Google ratings). But how do you make sure the right customers leave a review?One good indicator is their NPS score or the latest conversation rating. We have had tremendous success in generating 5-star reviews with a very simple Series: We ask customers that rated their last conversation positively and that are considered promoters based on their NPS to leave a review on a site of our choosing (we usually switch between the different review sites on a quarterly basis).

Hey Intercom Community, especially you - new to Intercom! 👋As I’ve been through being new at Intercom a long time ago, but seeing the rate of the development it has, sometimes I feel new again, so I feel like I can share some tips on how to get onboarded quickly and efficiently.  First of all, I’d like to start from this Community. At this stage of my Intercom journey it is the most important Intercom resource to me. Helping other people and solving their Intercom cases not just helps me sharpen my Intercom skills but also often spark some ideas on how I can achieve something with my own Intercom clients, that maybe I haven’t been thinking about it that way. 👍 I wish I have more time to be in here… Not that many service providers are putting so much effort in creating this kind of community like Intercom does.  Besides the Community, usual starting point is Intercom Academy. Besides getting the Intercom certification there, you can find additional courses that will help you learning

Hey Interconnecters! 👋 I’d like to share something more based on my Intercom experience, this time with focus on devs. I already kind of touched that in my previous post but feel like I really need to emphasize it a bit more.  Just in case you haven’t seen my previous post, make sure to check it out here: So, the devs. It’s not short for Devils but for Developers although some of my past colleagues might think the opposite 😋I’m old enough to know this one… Let’s get real - you’ll have dependency on devs if you want to make your Intercom project successful. In order to unleash all the power of Intercom, you’ll need to sync as much as possible data to it. And it’s almost impossible to do it without devs. Some 3rd party tools can help (i.e. Zapier, Automate.io, Make...) but eventually you will end up asking devs to help you out with something. I had many cases where I had to work with devs directly - and although I’m not a developer myself, I have enough technical background and experie