Video: Community Meetup Office Hours, May 11, 2026 | Duration: 3956s | Summary: Community Meetup Office Hours, May 11, 2026. Topics covered: Office Hours Welcome, Email Workflow Setup, Preventing Hallucinations, User Segmentation Guidance, Views and Filtering, Email Brand Configuration, User Login Workflows, Login Complexity, User Authentication Solutions, User Authorization Methods, User Verification Issues, Login ID Challenges, Login ID Recovery, Procedure Flow Continuity, Customer Introductions, CRM Migration Setup, Inbox Organization Setup, Backend Data Integration, Workflows and Integrations, Data Collection Planning, Document Sharing Plans, Community Wrap-Up, Empowering Support Teams, Closing Remarks, Closing Remarks | Chapters: Office Hours Welcome (2s), Email Workflow Setup (154s), Preventing Hallucinations (514s), User Segmentation Guidance (1185s), Views and Filtering (1442s), Email Brand Configuration (1679s), User Login Workflows (1893s), Login Complexity (1974s), User Authentication Solutions (2030s), User Authorization Methods (2106s), User Verification Issues (2219s), Login ID Challenges (2347s), Login ID Recovery (2420s), Procedure Flow Continuity (2558s), Customer Introductions (2750s), CRM Migration Setup (2806s), Inbox Organization Setup (2877s), Backend Data Integration (3010s), Workflows and Integrations (3165s), Data Collection Planning (3280s), Document Sharing Plans (3432s), Community Wrap-Up (3545s), Empowering Support Teams (3618s), Closing Remarks (3726s), Closing Remarks (3825s) Video: Community Meetup Office Hours, May 4, 2026 | Duration: 5508s | Summary: Community Meetup Office Hours, May 4, 2026. Topics covered: Welcome & Introductions, Office Hours Introduction, Q&A Setup, Q&A Setup, Workflow Escalation Rules, Testing Rule Conflicts, Testing Workspace Solutions, Test User Setup, Procedures and Simulations, Sales Agent Pricing, Presales Agent ROI, Q&A Management, Ticket Reassignment, Assignment Sweeps, Testing Conversation Attributes, Messenger Workflow Issues, Workflow Troubleshooting Tips, Browser Language Workflows, Workflow Troubleshooting, MCP Regional Availability, Unassigned Inbox Reporting, Customizing Fin Responses, Closing Remarks, Wrap-Up & Resources | Chapters: Welcome & Introductions (2s), Office Hours Introduction (38s), Q&A Setup (130s), Q&A Setup (219s), Workflow Escalation Rules (336s), Testing Rule Conflicts (417s), Testing Workspace Solutions (501s), Test User Setup (619s), Procedures and Simulations (731s), Sales Agent Pricing (889s), Presales Agent ROI (1083s), Q&A Management (1174s), Ticket Reassignment (1290s), Assignment Sweeps (1648s), Testing Conversation Attributes (1694s), Messenger Workflow Issues (2001s), Workflow Troubleshooting Tips (2114s), Browser Language Workflows (2504s), Workflow Troubleshooting (2618s), MCP Regional Availability (3060s), Unassigned Inbox Reporting (3289s), Customizing Fin Responses (3656s), Closing Remarks (4317s), Wrap-Up & Resources (4925s) Video: Community Meetup Office Hours, April 27, 2026 | Duration: 4388s | Summary: Community Meetup Office Hours, April 27, 2026. Topics covered: Office Hours Welcome, Welcome and Introductions, Team Check-In, Q&A Session, Fin Guidance Features, Audience Targeting Setup, Audience Targeting, User Audience Recognition, QA and Monitoring, Q&A Session, Audience Targeting Setup, Audience Targeting Setup, Audience Targeting Procedures, Troubleshooting Fin Recognition, Audience Targeting Procedures, Guidance Troubleshooting, Monitoring Fin Performance, Analyze and Monitor, Q&A Session, Fin Re-engagement Scenarios, Uploading Knowledge Data, Fin CLI Demo, Support Queue Management, Expert Resources, Support Contact Issues, Troubleshooting Support Scenarios, Escalation Rules Setup, Closing Remarks, Q&A Session, Data Upload Strategies, FIN CLI Setup, Open Q&A Session, Queue Management Solutions, Contact Support Issue, Closing and Feedback, Wrap Up | Chapters: Office Hours Welcome (5s), Welcome and Introductions (72s), Team Check-In (132s), Q&A Session (197s), Fin Guidance Features (340s), Audience Targeting Setup (488s), Audience Targeting (569s), User Audience Recognition (736s), QA and Monitoring (1231s), Q&A Session (1996s), Audience Targeting Setup (2029s), Audience Targeting Setup (2183s), Audience Targeting Procedures (2300s), Troubleshooting Fin Recognition (2512s), Audience Targeting Procedures (2594s), Guidance Troubleshooting (2721s), Monitoring Fin Performance (2998s), Analyze and Monitor (3215s), Q&A Session (3487s), Fin Re-engagement Scenarios (3665s), Uploading Knowledge Data (3877s), Fin CLI Demo (4181s), Support Queue Management (4463s), Expert Resources (4708s), Support Contact Issues (4802s), Troubleshooting Support Scenarios (5214s), Escalation Rules Setup (5692s), Closing Remarks (6037s), Q&A Session (6129s), Data Upload Strategies (6510s), FIN CLI Setup (6886s), Open Q&A Session (7077s), Queue Management Solutions (7140s), Contact Support Issue (7439s), Closing and Feedback (8205s) Video: Community Meetup Office Hours, April 20, 2026 | Duration: 3580s | Summary: Community Meetup Office Hours, April 20, 2026. Topics covered: Welcome & Introduction, Community Updates, Finn Workflow Testing, Reusable Workflow Setup, Workflow Branching Strategies, Escalation Workflows, Slack Notification Setup, Slack Workflow Troubleshooting, Troubleshooting Slack Routing, Troubleshooting Workflow Issues, Workflow Troubleshooting, Community Support Resources, Technical Difficulties, Q&A and Monitors, Intercom Features Discussion, Closing Recommendations, Wrap-Up and Next Steps | Chapters: Welcome & Introduction (4s), Community Updates (100s), Finn Workflow Testing (558s), Reusable Workflow Setup (1148s), Workflow Branching Strategies (1356s), Escalation Workflows (1544s), Slack Notification Setup (1597s), Slack Workflow Troubleshooting (1969s), Troubleshooting Slack Routing (2428s), Troubleshooting Workflow Issues (2488s), Workflow Troubleshooting (2571s), Community Support Resources (2708s), Technical Difficulties (2799s), Q&A and Monitors (2896s), Intercom Features Discussion (3025s), Closing Recommendations (3176s), Wrap-Up and Next Steps (3377s) Video: Community Meetup Office Hours, April 13, 2026 | Duration: 5252s | Summary: Community Meetup Office Hours, April 13, 2026 Video: Community Meetup Office Hours, April 6, 2026 | Duration: 3936s | Summary: Community Meetup Office Hours, April 6, 2026. Topics covered: Welcome & Introduction, Spam Filtering Setup, Attribute Taxonomy Strategy, Managing Multiple Brands, Email Forwarding Setup, Email Forwarding Setup, Q&A Session, Email Routing Logic, Procedures vs Guidance, Looping Through Organizations, Workspace Identification Solutions, Workarounds and Solutions, Workspace Integration Solutions, Custom Objects Discussion, Series Configuration, Closing Remarks, Closing and Community, Wrap-Up and Farewell | Chapters: Welcome & Introduction (4s), Spam Filtering Setup (150s), Attribute Taxonomy Strategy (311s), Managing Multiple Brands (472s), Email Forwarding Setup (743s), Email Forwarding Setup (1198s), Q&A Session (1456s), Email Routing Logic (1504s), Procedures vs Guidance (1735s), Looping Through Organizations (1952s), Workspace Identification Solutions (2124s), Workarounds and Solutions (2344s), Workspace Integration Solutions (2563s), Custom Objects Discussion (2721s), Series Configuration (3040s), Closing Remarks (3331s), Closing and Community (3503s), Wrap-Up and Farewell (3656s) Video: Community Meetup Office Hours, March 30, 2026 | Duration: 5132s | Summary: Community Meetup Office Hours, March 30, 2026 Video: Community Meetup Office Hours, March 23, 2026 | Duration: 5324s | Summary: Community Meetup Office Hours, March 23, 2026. Topics covered: Welcome to Office Hours, Introducing Office Hours, Reporting and Analytics, Batch Testing Strategies, Customizing Fin's Handling, Testing AI Responses, Intercom MCP Usage, Batch Testing Feedback, Intercom-Claude Integration, AI-Powered Data Insights, Troubleshooting Knowledge Sources, Troubleshooting Intercom Finn, Fin Targeting Issues, Audience Segmentation Challenges, HubSpot Integration Challenges, Enhancing Customer Information, Improving Finn's Responses, FinAI Guidance Refinement, Optimizing AI Guidance, Handling Support Escalations, Email Support Guidance, Email Integration Tips, Launching Finn Successfully | Chapters: Welcome to Office Hours (5s), Introducing Office Hours (38s), Reporting and Analytics (311s), Batch Testing Strategies (679s), Customizing Fins Handling (830s), Testing AI Responses (1167s), Intercom MCP Usage (1385s), Batch Testing Feedback (1596s), Intercom-Claude Integration (1773s), AI-Powered Data Insights (2079s), Troubleshooting Knowledge Sources (2726s), Troubleshooting Intercom Finn (2836s), Fin Targeting Issues (2934s), Audience Segmentation Challenges (3023s), HubSpot Integration Challenges (3091s), Enhancing Customer Information (3201s), Improving Finns Responses (3348s), FinAI Guidance Refinement (3458s), Optimizing AI Guidance (3594s), Handling Support Escalations (3766s), Email Support Guidance (3924s), Email Integration Tips (4131s), Launching Finn Successfully (4518s) Video: Community Meetup Office Hours, March 16, 2026 | Duration: 5192s | Summary: Community Meetup Office Hours, March 16, 2026. Topics covered: Welcome and Introduction, Claude and Intercom, Technical Difficulties Resolved, Company Data Management, Editing Company Information, InCycle for Data Management, User Permissions Issues, Editing Lead Data, Product Wishlist Navigation, Knowledge Base Writing, SLA Configuration Strategies, Community Support Discussion, FinAI Implementation Challenges, Iterating and Testing, Attribute Escalation Benefits, Iterating AI Guidance, Answering Customer Questions, Intercom Phone Features, Live Transcription Request, Wrapping Up Meeting, Improving Deflection Rates, Conclusion and Resources | Chapters: Welcome and Introduction (2s), Claude and Intercom (416s), Technical Difficulties Resolved (865s), Company Data Management (1067s), Editing Company Information (1298s), InCycle for Data Management (1386s), User Permissions Issues (1534s), Editing Lead Data (1658s), Product Wishlist Navigation (2041s), Knowledge Base Writing (2246s), SLA Configuration Strategies (2397s), Community Support Discussion (2539s), FinAI Implementation Challenges (2874s), Iterating and Testing (3534s), Attribute Escalation Benefits (3657s), Iterating AI Guidance (3771s), Answering Customer Questions (3960s), Intercom Phone Features (4033s), Live Transcription Request (4202s), Wrapping Up Meeting (4523s), Improving Deflection Rates (4540s), Conclusion and Resources (4826s) Video: Community Meetup Office Hours, March 09, 2026 | Duration: 3608s | Summary: Community Meetup Office Hours, March 09, 2026. Topics covered: Daylight Savings Confusion, Monday's Daylight Savings, UI Limitations Discussion, Community Feedback Discussion, Tracking SLA Times, MCP Server Discussion, Community Resources Overview, AI Integration Challenges, Concluding Thoughts | Chapters: Daylight Savings Confusion (1s), Mondays Daylight Savings (101s), UI Limitations Discussion (147s), Community Feedback Discussion (449s), Tracking SLA Times (590s), MCP Server Discussion (1043s), Community Resources Overview (1792s), AI Integration Challenges (2746s), Concluding Thoughts (3434s) Video: Community Meetup Office Hours, March 02, 2026 | Duration: 4424s | Summary: Community Meetup Office Hours, March 02, 2026. Topics covered: Community Q&A Introduction, Attributes and Reporting, Zendesk Integration Troubleshooting, Email Forwarding Troubleshooting, Troubleshooting Fin Integration, Troubleshooting Conversation Events, Configuring Fin Boxes, Fin Workflow Troubleshooting, Procedures and Resolution, Billing Integration Challenges, Troubleshooting App Conflicts, Leveraging Support Insights, Closing Remarks | Chapters: Community Q&A Introduction (5s), Attributes and Reporting (90s), Zendesk Integration Troubleshooting (558s), Email Forwarding Troubleshooting (1356s), Troubleshooting Fin Integration (1951s), Troubleshooting Conversation Events (2125s), Configuring Fin Boxes (2240s), Fin Workflow Troubleshooting (2430s), Procedures and Resolution (3088s), Billing Integration Challenges (3492s), Troubleshooting App Conflicts (3657s), Leveraging Support Insights (4036s), Closing Remarks (4336s) Video: Community Meetup Office Hours, February 23, 2026 | Duration: 5272s | Summary: Community Meetup Office Hours, February 23, 2026. Topics covered: Welcome to Office Hours, Welcome and Questions, Intercom Goals Discussion, CX Score Explained, Tags vs Attributes, Conversation Management Strategies, Tickets vs Conversations, Conversations vs. Tickets, Ticket vs Conversation, FIN Procedures Overview, Procedures and Simulations, Procedures and Simulations, Procedures vs. Escalation, Procedures vs. Guidance, Report Discrepancy Discussion, Troubleshooting SLA Reports, Gathering Customer Feedback, Refining Feedback Attributes, Feature Request Analysis, Sharing Cloud Artifacts, Customer Trust Issues, Concluding Data Analysis, Visual Report Generation | Chapters: Welcome to Office Hours (5s), Welcome and Questions (69s), Intercom Goals Discussion (157s), CX Score Explained (222s), Tags vs Attributes (556s), Conversation Management Strategies (921s), Tickets vs Conversations (1080s), Conversations vs. Tickets (1187s), Ticket vs Conversation (1301s), FIN Procedures Overview (1464s), Procedures and Simulations (1840s), Procedures and Simulations (2061s), Procedures vs. Escalation (2519s), Procedures vs. Guidance (2678s), Report Discrepancy Discussion (2845s), Troubleshooting SLA Reports (2989s), Gathering Customer Feedback (3209s), Refining Feedback Attributes (3367s), Feature Request Analysis (3806s), Sharing Cloud Artifacts (4984s), Customer Trust Issues (5057s), Concluding Data Analysis (5131s), Visual Report Generation (5237s) Video: Intercom Office Hours, February 11, 2026 | Duration: 3472s | Summary: Intercom Office Hours, February 11, 2026. Topics covered: Welcome and Introductions, Skipping Finn Automation, Help Center Focus, Support Training Tools, Customer Support Challenges, Intercom Conversation Management, Clarifying User Issues, CSAT Interface Updates, Maximizing Intercom Engagement, Bulk Tagging Discussion, Customizing Intercom Help Centers, Booking Priority System, Client Survey Integration, Small Wins, Big Impact, Closing Remarks, Feedback and Validation, Concluding Audience Discussion | Chapters: Welcome and Introductions (6s), Skipping Finn Automation (133s), Help Center Focus (315s), Support Training Tools (647s), Customer Support Challenges (1061s), Intercom Conversation Management (1203s), Clarifying User Issues (1297s), CSAT Interface Updates (1367s), Maximizing Intercom Engagement (1446s), Bulk Tagging Discussion (1699s), Customizing Intercom Help Centers (1860s), Booking Priority System (2597s), Client Survey Integration (2698s), Small Wins, Big Impact (2794s), Closing Remarks (2937s), Feedback and Validation (2965s), Concluding Audience Discussion (3028s) Video: Community Meetup Office Hours, February 9, 2026 | Duration: 3864s | Summary: Community Meetup Office Hours, February 9, 2026. Topics covered: Welcome and Introduction, Guidance vs. Snippets, AI Implementation Challenges, AI in Support, Understanding User Preferences, Email Workflow Challenges, Email Setup Strategies, Optimizing AI Responses, Wrapping Up Q&A, Concluding Q&A Session | Chapters: Welcome and Introduction (4s), Guidance vs. Snippets (99s), AI Implementation Challenges (367s), AI in Support (691s), Understanding User Preferences (1251s), Email Workflow Challenges (1280s), Email Setup Strategies (1469s), Optimizing AI Responses (1715s), Wrapping Up Q&A (2184s), Concluding Q&A Session (3443s) Video: Community Meetup Office Hours, January 26, 2026 | Duration: 3908s | Summary: Community Meetup Office Hours, January 26, 2026 Video: Community Meetup Office Hours, January 19, 2026 | Duration: 4724s | Summary: Community Meetup Office Hours, January 19, 2026. Topics covered: Welcome and Introductions, Q&A and Troubleshooting, Troubleshooting AI Dashboard, CX Score Issue, Technical Support Discussion, AI Browser Options, AI Privacy Considerations, Audio Troubleshooting Session, Customizing Product Categories, Intercom Feature Suggestions, Survey and Feedback, Knowledge Base Structure, Understanding Customer Interactions, Copilot Usage Challenges, Intercom Team Challenges, Task Design Process, Fintask API Integration, Answering User Questions, Managing Global Issues, Closing Thoughts | Chapters: Welcome and Introductions (4s), Q&A and Troubleshooting (102s), Troubleshooting AI Dashboard (285s), CX Score Issue (352s), Technical Support Discussion (501s), AI Browser Options (566s), AI Privacy Considerations (700s), Audio Troubleshooting Session (869s), Customizing Product Categories (2215s), Intercom Feature Suggestions (2757s), Survey and Feedback (2893s), Knowledge Base Structure (2970s), Understanding Customer Interactions (3187s), Copilot Usage Challenges (3300s), Intercom Team Challenges (3455s), Task Design Process (3595s), Fintask API Integration (3727s), Answering User Questions (3851s), Managing Global Issues (3982s), Closing Thoughts (4325s) Video: Community Meetup Office Hours, January 12, 2026 | Duration: 5208s | Summary: Community Meetup Office Hours, January 12, 2026. Topics covered: Welcome and Introduction, Escalation Rules Explained, Custom System Integration, Connecting External Systems, Handling Sensitive Data, Data Connector Challenges, Addressing User Questions, Q&A on Integrations, Intercom Integration Discussion, Exploring Copilot Features, Browser-Based Support Solutions, AI Browser Integration, Fin Attributes Setup, Attribute Usage Benefits, Concluding Gratitude, Addressing Bailey's Questions, CX Score Concerns, Implementing Intake Forms, Conclusion and Recap, Improving CX Score | Chapters: Welcome and Introduction (4s), Escalation Rules Explained (243s), Custom System Integration (538s), Connecting External Systems (718s), Handling Sensitive Data (1016s), Data Connector Challenges (1293s), Addressing User Questions (2069s), Q&A on Integrations (2617s), Intercom Integration Discussion (3277s), Exploring Copilot Features (3349s), Browser-Based Support Solutions (3440s), AI Browser Integration (3574s), Fin Attributes Setup (3732s), Attribute Usage Benefits (3950s), Concluding Gratitude (4131s), Addressing Baileys Questions (4137s), CX Score Concerns (4162s), Implementing Intake Forms (4394s), Conclusion and Recap (4693s), Improving CX Score (5192s) Video: Community Meetup Office Hours, January 5, 2026 | Duration: 4468s | Summary: Community Meetup Office Hours, January 5, 2026. Topics covered: Welcome and Introduction, Addressing SLA Reporting, SLA Reporting Challenges, Intercom SLA Reporting, Intercom Reporting Features, Addressing Customer Inquiries, Ticketing System Necessity, Integrations and Apps, AI Help Center, AI-Powered Customer Onboarding, Intercom AI Features, Simulations and Tasks, Updating Image Responses, Image Optimization Tips, Image Tips Newsletter, Automation and Training, Automating Inactive Conversations | Chapters: Welcome and Introduction (4s), Addressing SLA Reporting (89s), SLA Reporting Challenges (237s), Intercom SLA Reporting (373s), Intercom Reporting Features (553s), Addressing Customer Inquiries (828s), Ticketing System Necessity (943s), Integrations and Apps (1213s), AI Help Center (1729s), AI-Powered Customer Onboarding (1966s), Intercom AI Features (2270s), Simulations and Tasks (2569s), Updating Image Responses (2736s), Image Optimization Tips (2811s), Image Tips Newsletter (2962s), Automation and Training (3078s), Automating Inactive Conversations (3302s) Video: Community Meetup Office Hours, December 29, 2025 | Duration: 2684s | Summary: Community Meetup Office Hours, December 29, 2025. Topics covered: Welcome to Office Hours, Holiday Message Setup, Managing Holiday Notifications, Tracking Fin Performance, Custom Report Scheduling, CX Score Improvements, Intercom Community Benefits, Feedback and Automation, Improving Fin's Performance, Customizing Fin Reports, Wrapping Up Discussion, Wrapping Up Session, Concluding Community Reflections | Chapters: Welcome to Office Hours (5s), Holiday Message Setup (95s), Managing Holiday Notifications (434s), Tracking Fin Performance (669s), Custom Report Scheduling (774s), CX Score Improvements (1278s), Intercom Community Benefits (1348s), Feedback and Automation (1437s), Improving Fins Performance (1610s), Customizing Fin Reports (1657s), Wrapping Up Discussion (1882s), Wrapping Up Session (2001s), Concluding Community Reflections (2129s) Video: Community Meetup Office Hours, December 22, 2025 | Duration: 3628s | Summary: Community Meetup Office Hours, December 22, 2025. Topics covered: Welcoming the Community, VIN Capabilities Discussion, Snippets vs Guidance, Office Hours Management, Intercom Holiday Hours, AI Image Understanding, Intercom Integration Challenges, Intercom Integration Solutions, Interactive Demo Challenges, Intercom Integration Challenges, Embedding Integration Discussion, Exploring Integration Tools, Mobile App Automation, Community Recognition Highlights | Chapters: Welcoming the Community (7s), VIN Capabilities Discussion (101s), Snippets vs Guidance (166s), Office Hours Management (355s), Intercom Holiday Hours (753s), AI Image Understanding (1355s), Intercom Integration Challenges (1434s), Intercom Integration Solutions (1530s), Interactive Demo Challenges (1669s), Intercom Integration Challenges (1802s), Embedding Integration Discussion (1912s), Exploring Integration Tools (2108s), Mobile App Automation (2431s), Community Recognition Highlights (2593s) Video: Community Meetup Office Hours December 15, 2025 | Duration: 5308s | Summary: Community Meetup Office Hours December 15, 2025. Topics covered: Welcome and Introduction, Navigating Inbox Features, Team Inbox Features, Workflow Consolidation Strategies, Workflow Channel Considerations, Improving Involvement Rates, Intercom MCP Discussion, Intercom Suggestions Feature, WhatsApp Integration Limitations, Sharing AI Prompts, Sharing Guidance Pieces, Effective Prompting Techniques, Tracker Ticket Management, Ticket Management Systems, Q&A Session Continues, User Communication Challenges, Context Clarification Guidance, Context and Clarification, Guidance Usage Tips, AI-Powered Customer Support, AI Task Optimization, Improving AI Escalation, Handover Process Optimization, Configuring FIN Deployment, Contextual Language Analysis | Chapters: Welcome and Introduction (6s), Navigating Inbox Features (151s), Team Inbox Features (330s), Workflow Consolidation Strategies (422s), Workflow Channel Considerations (545s), Improving Involvement Rates (829s), Intercom MCP Discussion (1077s), Intercom Suggestions Feature (1284s), WhatsApp Integration Limitations (1538s), Sharing AI Prompts (1893s), Sharing Guidance Pieces (2025s), Effective Prompting Techniques (2136s), Tracker Ticket Management (2269s), Ticket Management Systems (2374s), Q&A Session Continues (2537s), User Communication Challenges (2853s), Context Clarification Guidance (2923s), Context and Clarification (2985s), Guidance Usage Tips (3150s), AI-Powered Customer Support (3406s), AI Task Optimization (3764s), Improving AI Escalation (3990s), Handover Process Optimization (4489s), Configuring FIN Deployment (4653s), Contextual Language Analysis (4694s) Video: Community Meetup Office Hours, December 8, 2025 | Summary: Community Meetup Office Hours, December 8, 2025. Topics covered: Welcome and Introduction, Welcome Back Session, Audience Filtering Challenges, Content Management Challenges, Troubleshooting Community Issues, Knowledge Management Challenges, Data Connector Setup, Data Connector Strategies, Task Implementation Example, Managing Delivery Changes, Data Connector Alternatives, API Configuration Strategies, Checking Workflow Effectiveness, Data Connector Insights, Simulations Functionality Explained, Simulations and Feedback, User Classification Challenges, Audiences for Personalization, Audience Configuration Tips, Managing Snippets Effectively, Managing Content Snippets, Exporting Knowledge Articles, Concluding Q&A Session, Wrapping Up Q&A, Buttons vs. Fins, Future of Intercom, AI-Driven Routing Strategies, Workflow Routing Demonstration, Workflow Customization Tips, Closing Remarks | Chapters: Welcome and Introduction (3s), Welcome Back Session (163s), Audience Filtering Challenges (248s), Content Management Challenges (527s), Troubleshooting Community Issues (594s), Knowledge Management Challenges (667s), Data Connector Setup (833s), Data Connector Strategies (999s), Task Implementation Example (1102s), Managing Delivery Changes (1215s), Data Connector Alternatives (1321s), API Configuration Strategies (1420s), Checking Workflow Effectiveness (1533s), Data Connector Insights (1668s), Simulations Functionality Explained (1795s), Simulations and Feedback (1924s), User Classification Challenges (2107s), Audiences for Personalization (2249s), Audience Configuration Tips (2438s), Managing Snippets Effectively (2675s), Managing Content Snippets (2767s), Exporting Knowledge Articles (2895s), Concluding Q&A Session (3818s), Wrapping Up Q&A (3857s), Buttons vs. Fins (3922s), Future of Intercom (4086s), AI-Driven Routing Strategies (4198s), Workflow Routing Demonstration (4303s), Workflow Customization Tips (4571s) Video: Community Meetup Office Hours, December 1, 2025 | Summary: Community Meetup Office Hours, December 1, 2025. Topics covered: Intercom Community Welcome, Introducing Expert Panelists, Community Experts Introduction, Introducing Community Experts, Content Management Strategies, Intercom Community Benefits, Community Learning Strategies, Optimizing Finn Performance, Optimizing with MCP, Optimizing Conversation Volume, Intercom Product Updates, Simulation Benefits Discussed, Q&A and Privacy, Naming AI Assistants, Wrapping Up Discussion, Community and Integration, Conclusion and Resources | Chapters: Intercom Community Welcome (5s), Introducing Expert Panelists (72s), Community Experts Introduction (216s), Introducing Community Experts (331s), Content Management Strategies (461s), Intercom Community Benefits (1132s), Community Learning Strategies (1401s), Optimizing Finn Performance (1563s), Optimizing with MCP (1907s), Optimizing Conversation Volume (2162s), Intercom Product Updates (2223s), Simulation Benefits Discussed (2657s), Q&A and Privacy (2719s), Naming AI Assistants (2801s), Wrapping Up Discussion (2928s), Community and Integration (3043s) Video: Community Meetup Office Hours - Procedures & Data Connectors | Duration: 3960s | Summary: Community Meetup Office Hours - Procedures & Data Connectors. Topics covered: Welcome and Introduction, Team Introductions, Participant Introductions, Breakout Room Setup, Q&A Session, Troubleshooting Procedures, Procedures Overview Demo, Data Connector Integrations, Data Connector Setup, FIN CLI Demo, API Integration Examples, Shopify Connector Demo, Shopify Data Connectors, Shopify & KYC Integration, Dashboards and Reporting, Dashboard Insights, Q&A Session Management, Q&A Use Cases, Session Wrap-Up, Closing Remarks | Chapters: Welcome and Introduction (6s), Team Introductions (73s), Participant Introductions (193s), Breakout Room Setup (266s), Q&A Session (362s), Troubleshooting Procedures (558s), Procedures Overview Demo (778s), Data Connector Integrations (1143s), Data Connector Setup (1637s), FIN CLI Demo (1780s), API Integration Examples (1886s), Shopify Connector Demo (2050s), Shopify Data Connectors (2171s), Shopify & KYC Integration (2457s), Dashboards and Reporting (2722s), Dashboard Insights (3039s), Q&A Session Management (3214s), Q&A Use Cases (3342s), Session Wrap-Up (3452s), Closing Remarks (3748s)

Transcript for "Community Meetup Office Hours, March 23, 2026": Hey, everyone. What's up? It's the, office hours as usual every Monday. We got a number of people already joining. Feel free to join, drop in the chat. Some welcomes. Where are you coming from? What's up? Joel from Florida. How's it going? Good to have you back, Joel. My name is Nathan, obviously, as you can see down here. I'm one of the, top community experts in the intercom community, and I'm here to answer questions, have a chat, talk about random things to do with intercom. What we're not gonna do today is we're not gonna do some promotions and pitches and all that stuff. We're just here to answer questions. So what's up, Edward, from UK? My week has been it's like is it Monday already? Because it feels like the week's already been busy. Jess, what's up? Welcome. New to Intercom. Yeah. It's always good to have new people from to Intercom here. Lillian from Calgary. Alright. What's up? I'm from Canada originally too. I'm over in Paraguay right now in South America. So but, yeah, nice to have some Canadians with us. Yeah. If you got any questions, start dropping them in the q and a or drop them in the chat, and we'll move them over there. What's everybody doing today? How's the how's the inbox going this morning? Are you how many people are, like, actually manning the inbox while you're actually here on the call too? Are you, monitoring the inbox and the and the and the office hours at the same time? Jess. Alright. Cool. Cool. Nice. Oh, New York. Alright. Cool. California, New York. First time attendee, Marcel. Appreciate it. Welcome. I'm gonna say, is this so Silvana? Is that how I say your name? I'm so curious now. It's a new spelling of a name for me, so that's cool. I love I always like learning new names and new pronunciations even though I don't always do the best job of guessing. Okay. I guess I did alright this time. So what's everybody got on their mind today? What are you thinking about? What do you what do you wanna talk about? Feel free to drop in the chat. Everybody's also willing to, you know, keep at introducing yourself. Let us know where you're coming from too. I know new new people are joining. So we're here every Monday and basically just answer your questions. So if I can't answer it, I'll try to, you know, link to something that does or, get someone else to jump in and help us. And, you might even be able to answer some other, people's questions as well. That's what we've, noticed a lot here. We are we're also we're all experts in some way to some part of Intercom, the parts we're learning, the parts we know. Patrice, welcome. New Jersey. Alright. Another first time attendee. This is cool. We got a number of people here for the first time. So yeah. So we're like, again, if it's your first time here, Garrett, first time. Alright. Yeah. Yeah. Yeah. We got a lot of first timers. So I hope you have lots of questions. We got an hour. Sometimes I stay a little longer if we can't answer them all. My goal is to answer all the questions today and at least give you some insights on things you can work with and do. So I don't I'm not from Intercom directly. I'm I'm a community expert. I just help other other teams and companies, using Intercom. So you can ask any questions, and I'll give you real, like, real world experience on that. I have some insight, obviously, from the Intercom team directly. I, you know, I connect with them and get info and, know, ask questions. But, most of my knowledge comes from just using Intercom myself. I worked for a start up for three years where I moved them from Zendesk over to Intercom and then realized, like, when I left that start up, I was like, I have all this knowledge on Intercom. I'll just keep helping people and answering questions. And then one day, I realized I could just do this as what I do. So that's what I've been doing now for the last, several years, maybe three or four years. And, it's a lot of fun helping people. The one thing I always say, and a number of you that are new here probably, you know, will will hear me say this a lot, along the way, but you already have a full time job. And then you're now have this other full time job trying to learn intercom. And so, what what we try to do as community experts is help you skip that learning curve as much as possible by answering questions and helping, but also we can come alongside you and, you know, help you in that way. So not to overpromote this, but there's a meet the experts button up here. You can see my information on myself and a bunch of other experts, Julian, Connor, Milan, Nur, some others, if you need some help and look more in-depth help. But Diana, who's community lead, she's not here today. She's she's out on for personal stuff, but she said she really we worked together to make this possible so we could answer your questions because we felt like, you know, the webinars and all that stuff are great, but a lot of times, we're squeezing people are squeezing in, like, that little last few minutes, like, trying to answer all your questions and, you know, always getting answered. So so that's why we're here, just like a whole hour of just answering your questions. So but so far, I don't see any questions in the in the q and a. So do you guys have questions? Do you do you wanna just talk? Is there anything you wanna talk about? Give me some topics if you don't have specific questions, so we can maybe dig into stuff. But I'm hoping you got some questions since you're here today. Okay. Jess is starting us off. Okay. Perfect. How can I get a total number of let me see? Can I pin this? I haven't been able to pin this to the board very easily lately for some reason. So if you ask a question, I'm just gonna copy and paste it over probably to the q and a if I can't do it for some reason. Okay. And we've got another one in the q and a already too, so that's cool. Okay. So I'm gonna go with Jess's. I'm just gonna read it out loud. Actually, let me yeah. Okay. I'll just read it out loud because I can't pin it to the thing for some reason. Usually, I can pin it to the board. How can I get a a total number of conversations closed of each teammate under my conversation close report? I would like to be able to I like I like it to be directly on my report, not a CSV. Interesting. Okay. Well, you know what? We got a few questions getting in the chat, but in the meantime, why don't I, go ahead and try to show you how to do this? To be honest, I have made a chart a report like these kind of reports before, but I'm just gonna do it on the fly, which will be more difficult, I think. So let's go I'm just gonna go to I'll go to I don't have a lot of team members in in my own intercom, but I'm gonna go ahead and use my own intercom because I don't wanna put up any client stuff on the screen. You guys can see my screen okay? You're gonna see my terrible ratings on my own product with my own response time because I am not a good, support agent. I'm a much better, person at configuring the system than actually running the system myself, and that's just a honest answer. So, so you wanna do, like, a team in inbox performance type thing or a teammate performance. Like, this kind of this kind of setup is already built in, but you want something for a very specific teammate. Right? Like so you could actually just pick a specific teammate to run your like, to to configure the report using a filter, something like this. You know, the other thing that I would say that you can do you wanted to do you see closed conversations. Right? So if you see your conversation report here and you have new conversations and there's, closed Just trying to see if I can add a widget. So we can add a chart for I don't know if we have, but we have one specifically closed conversations. So, yeah, so we can just do closed conversations here. There's four. I don't have a lot of conversations in this inbox, but you could just put a filter on it to say who it's for. Right? So team teammate assigned is, and then pick your teammate. And then you could update that chart so you could have it. Essentially, I have one here that has, like, no closed conversations for me in the last period of time that we're looking at. Does something like that, like, help you, Jess? I know it's, like, kind of, like, on the fly. Might not be so clear, but let me know if that if that helps. Like, just being able to, like, make a a like, the big thing is being able to know that you can add filters to any chart so you can pick the teammate assigned. So if you get a chart you like, you can even do it at the top here. You know, just add the teammate assigned to it. And then if you want, you can just, you can just rate it. I mean I mean, you could just save it and give it a name, like, by so and so, closed conversations by so and so. The other thing I was gonna show you I mean, say to you is, like, if you haven't used the intercom MCP yet, I would highly recommend using that. Cool. Did that did that help, Jess? Sorry. I'm just trying to figure out okay. Yes. Closed in total past days. Yeah. If you just set up a closed chart with total past seven days and then put the filter on for that particular person, then you could see all those for that person, if that helps. I'm gonna ask Terry. Is it Terry? What's up, Terry, by the way? Can you move your question to the q and a for me? Like, just copy and paste that over there just because I for some reason, I can't pin anything to the board today from the chat. So anybody who's got a question, if you can drop in the q and a tab off to the right side there, that will be really helpful. Perfect. Thank you so much. Okay. So let's get to Savannah's question here. And then oh, great. Okay. Cool. This is a good one. Alright. So I'm sure a lot of people will have questions like this if you're first time here. So here's Savannah's question. And I'll I'll also, if you wanna come on stage and talk about this, feel free to raise your hand in the app, and then I can bring you up on stage, we can talk about it a bit more. So as far as launching Finns, you're working on the process of getting Fin Live. What was your approach to it? This is a, you know, this is a big question, obviously. But, like, the short answer, I would say, is make sure, like, one, loading your content, making sure your content is current as possible, and then run some tests as much as possible, like the batch testing functionality if you haven't used that yet. If you're using any procedures, you've got the added bonus of being able to use simulations, which means you can actually simulate, like, the whole conversation, which is way, way better than just, like, batch testing, but batch testing can help a lot. Have you done any batch testing or anything yet, Savannah? And if you wanna jump on stage and talk about this instead, that would be good. Okay, Jessica. Above the zero, I just saw that you said that it was zero for you as well. That's interesting. So if you're having a if you're getting a zero a zero response on that and there's an issue with the chart or something, definitely need to mention that to the intercom team, but maybe we can use a real example for that. I'll show you the intercom MCP shortly too because I think, honestly, reporting if you're making a chart or reporting or looking into stuff, intercom MCP is a better choice. So we'll come back to that again. I'll show you in a minute. Okay. Savannah, raise their hands, and let's get I'm I'm assuming. Like okay. Savannah, sorry. All bad, bro. My bad. My bad. My bad. Okay. good. I was like. I just like, this name name is an assumption on my part, so that's my bad. Yeah. No. All. good. You worries. get that a lot, I guess. You get. that a lot, do. I'm guessing. I I appreciate it, though. bad. I I try not to make that assumption, and I just did it at the last second. My my apologies. Nope. Yeah. You're good. go. You're good. So to your question, no. I haven't used batch testing yet. We haven't used the procedures as well either. I think right now, we're. we're, like, manually reviewing a lot of our chats just to kinda, like, get an idea of Sure. what type of queries are coming in and then kinda. going from there. I guess, like, my question is, like, do you just because I I hear a lot people just like, oh, we're just we just release Finn and just let it go crazy and see what comes back. with it. Yep. We're we're more in the financial kinda aspect, Yeah. so we we wanna that. yeah. You can't take that really. release that. Yeah. So. Mhmm. is that approach better, or is, like, the slow and steady kinda thing, like yeah. I think I think you got two things I would say with this. Like, one, if you're in that kind of space, like, you need to know your content's working. So, definitely, you gotta do the batch testing, in my opinion. So you you might have seen at the at the webinar, I think it was, like, Pioneer, the big event. There was, like, this back and forth between two people. There was one who was like, we just let it loose and see what happened, and then we may iterate on the fly. And the other person was like, no way. We never do that. Like, we we have. to do it strategically. You know? So I think, like, you're gonna have two camps that are always like, there's always the people that just let it fly and let's see what happens, and there's the people that are like, no. We gotta test everything thoroughly. I think in your case, like, especially in the financial space, there's a lot of sensitive spaces where you don't wanna do that. Like, you wanna test it out. What you can do, though there's two different things that I would recommend here. One is obviously doing a lot of batch testing. I don't if you if you're if you're familiar with that. You do the batch testing. You can basically generate questions from your conversations. So it'll pull in all these questions based on actual content in your inbox, and then you can just kind of go through it. And we can pull it up on the screen in a minute and show that. But, basically, you go through and you mark it, like, poor or good or whatever, and then you can, like, put why. Then you can go back and make some changes, come back and run it again and see if you've gotten, you know, better responses. So that's the first step. The second thing is, like, if there's an area where you find it's doing really well with those things, but then you can basically go into Fin and tell Fin, don't handle it like, only handle these kind of questions or don't handle this kind of stuff. And, like, I'll give you one example from a client I'm working with right now, actually. So we have a client right now who all they want is they just want Finn to handle product questions. And if it's not a product question, they don't wanna answer it. Right? They don't want Finn to do anything with it. So what we did was we put in Finn attributes that don't know if you're familiar with Finn attributes. It used to be called AI categorization. You basically set up all your attributes so that that FIN knows what kind of conversation it is. Like, oh, this is a product question, or this is, like, troubleshooting, or this is support, or this whatever. And then you can basically go into FIN into the escalation and say, you know, you know, if it's not a product question, escalate it to the team. You know? And if it is a product question, then Fin can handle it. So you can you can get very granular with, Mhmm. like, what Fin's allowed to handle, and you can also, like, kinda route everything but the things that you want. And so you can start with one category that you feel like, okay. Fin's handling these these questions very well. K. Let it handle those ones and then start seeing how it works and then start working on the other ones. The other thing you can do is just, like, get very clear on the stuff that you like, if you if you think, like, hey. I want it to handle everything, but I don't want it to handle these particular things. You could go the other way around, which is, like, let it handle everything, but, hey. These this type of question, this type of question, no. That gets routed to the to the team immediately. And so, I think that's kinda your best approach when it comes to financial stuff. Just figure out, like, is it mostly able to handle these questions well, and I just wanna route a couple things out that I don't want, you know, it touching? Or is it, like, most of the stuff should get routed to the team and just try to really focus on this one area that you have high ROI on and, like, good accuracy with? And I think, like, procedures I I honestly would suggest procedures, obviously, in the in the financial space because you could actually decide the order that it checks things and does things, and that might be really useful. Like, hey. If the if the client's on this particular plan, you know, and their their qual like, certain qualities are this, you know, then use this procedure and then follow these steps. You know? And it's like, if they're not on this plan, they're not on this thing, then don't then then it'll follow different steps. Right? So. what what it so you haven't done any testing yet, you're not exactly sure how it's actually gonna work at all at the moment. Right? No. And I think I think our goal was to get procedures up and running, but that really just. depends on our our platforms and engineering team on how quick they they are moving. Yep. So. it's kinda. scrappy. at. the moment right now just with, I guess, the batch. testing and the manual testing. Yep. Oh, You can do some you can do some procedures without data connectors as well. Like, you can try running them and seeing how they work and get the idea of, like, what what they should be doing before you get there, and that gives you the ability to run the simulations and see how Fin answers it. You know, I can I I can share my screen for a second and just show you a couple things that I think would be useful here? And anybody else listening in too, like, you've got feedback, feel free to drop in the chat for for Savanna here too because that's that's important, you know, on not just me, but other people's opinions here. Let me go into I just wanna go into one of these client accounts that I can actually show you. Okay. Duplex here. So we have we have, like so batch testing. We did quite a bit of batch testing, although there wasn't a lot of batch testing functionality at the time. I actually also do have a beta of some simulations on the top level, which I'm not sure if they're gonna roll out to everyone, but I really think, like, it's useful. Like, if you can run a simulation at this top level yeah. They don't have it they don't have it now, but they did have a beta where I could run simulations at the top level. But the the batch testing is good, and, also, you can run preview in preview, you can run a bunch of that stuff as well. But the key thing here, if you're starting out, is make a new make a new question group and generate it from the inbox, and you can do it by different topics. I'd highly recommend doing it by topics. So Finn has picked up on a lot of different topics here. And so I can say, oh, like, run run a a batch of questions, basically, on color formats. Like, this is a printing company, Jukebox. They do business cards and and and stickers and all kinds of flyers and that kind of stuff. So I can, like, pick one of my top topics. Maybe that's not the best one. Maybe shipping is better. So I can say, like, okay. Let's do, like, shipping questions. And then I run this, and, basically, it generates from the inbox based on the topic a bunch of questions related to shipping. And then you can go through and see, like, how well is Finn doing this out of the box. Right? And you can you can tweak it, investigate it, and then you can run this all over again. So if anybody's watching, you have questions about this while we're doing it, just let me know. And, obviously, of course, if you have some questions about it, just just ask me. But so it's gonna generate this. It generates about 50 questions usually, and then you can you can go through and, like, see how relevant the questions are, delete them if you want. And then you can also send the team to go look at it and give their feedback. The one thing I've been asking the the the the Intercom team for is a way to, I'm gonna flip to a different category while this is generating. So I have a I always have, like, a random question category where I just add, like, ones as I need them. So this is an example of how it works. Like, I'm like, what are your office hours? We're trying to figure out if the office hours information was correct. And so it generated the office hours, and I can see which content is used to create that. So if I need to troubleshoot it, I can look in there, and I can look at which guidance it's using. And then this is where you can say, like, oh, it's good. It's acceptable. It's poor. And if you click on poor, it asks you why is it poor. You know? Didn't use the correct content. Didn't clarify the customer's question. You know, used the used the content incorrectly, tone wasn't right, whatever. You pick what it is. So if I say, like, oh, didn't use the correct content, then it's gonna say, okay. Improve this answer. And then it takes me to the improve the answer workflow where I can, like, either create a snippet or add update some content that it tried to that it tried to use, or I can like, it's got a couple other different pieces of content. And then you can this is kinda walking through improving it. And then you can come back, and then you click this button, and you run it again, basically. And then it'll check that after you've made your changes, and then you see how you're you know, see how the response is. And then once you've gone through, like, some of your top questions in a specific category and you kinda feel like they're all, you know, good good answers, then you can feel more comfortable in turning it on for a specific category of of information. You know? And then I can go back now to see this the shipping question should be there. Yeah. So now we have, it doesn't say yeah. 50 questions. So we have 50 shipping questions, and we can click on each one of these and see the answer. In this case, you can see that it actually triggered one of my tasks, so it doesn't show the details. This one, you know, triggered an answer. This one is it is asking for an order number. So if you run into this kind of problem where you have some sort of system that's asking for something in advance, then what you can do is you can run it as a specific user or a specific audience member. So, like, if you if you find if anybody's done this before and you have some tasks in place, and you need something to run that, then you just switch it to you can just switch it to somebody who, in this case, I don't have it set up properly, but I can just pick any user. I think I can have myself in here, actually. So I can switch it to myself, and then I should it should have the information to rerun the test as myself, and then it should give me the information that I need. So you can test it as a real user and not just as, like, a generic profile where where, like, if you need actual data from the customer to get the answer, you can get it. So to see, now it answered it answered my question differently. Actually, I don't if that's the same question I was on. But, yeah, like, it answered my question differently because it recognized my tracking number. But this this one here yeah. Anyways, long story short, you can do that. I actually have, in my other client setup I think I have let's see. You can run it with a serial number. And, actually, in the preview area now, you can also do this. So if you're I'm gonna come back and see if anybody's got questions in the chat in a second here. See, on the on the preview side, you can ask questions here, and that allows you to ask, like, follow-up questions and not just batch testing. But keep in mind that anything you put in the preview section here goes into the inbox as. a preview user. So you just gotta alert your team that they don't, like, think it's an actual real conversation and start replying to it. But here, you can do, audiences. So you can set up an audience of people, or you can set up, a user as well. So you can do the same thing I was talking about, before. So you can, like, see how does this work with a specific customer. But, again, make sure you let your team know first because as these go into the chat, you know, some of them some people won't notice because the you've kind of been trained to, like, ignore the name when it's a chat person because it's like, Yeah. you know, pink giraffe said what or whatever. Right? So it's like, you sometimes don't look at the name and just start answering the questions. So did. did any questions come up for you while we're talking about that? Or No. I mean, I I don't wanna take up too much of time. So I know everyone has. other questions. I guess my last thing is. just, like, when you generate the batch questions, is it only 50. questions per folder, or or how does that work? Yeah. It's it's 50 questions per group, but you can have as many groups. I think you can have as many groups as you want. I haven't ever run out of groups before. I kinda wish it generated less questions because, like, sometimes I wanna add my own questions. I have to delete some to add some to the group. That makes sense. would be nice if you could say, like, generate 10 questions or something instead of generating 50, you know, all the. time. But that's how it's set up right now. The other. thing you can do is to start out with, if you want, you can just generate conversations from the inbox and don't pick a topic, and then it'll just kinda it'll try to pick the top questions across all the topics. And then you can start there and then work your way through it and then maybe do topics for the ones that you care about the most. Right. But, yeah, that's that's the way to start. And the other thing I would say too, which is kinda going back to Jess' comments, so maybe I can show you both of these. Like so the other thing, if you haven't used the Claude with the Intercom MCP yet, I would that's, like, one of the places I start now when I'm working with clients on Fin. So let me share my screen again, and I'll show you this. So if I go into Claude I'm just gonna start oh, wait. I'm in Claude. You can do it in Claude Cowork, or you can do it in in chat. Either way, I'm just gonna I don't know which inbox I'm connected to for clients right now, but I'm just gonna give you an example. So if you're getting started with launching Fin, you can actually go in here and say, like, you know, if you wanna get a list of top questions, you can say you could just ask it the topic. Like, what are the top the top questions asked this month? I'm just doing it general because I don't know which inbox I'm actually in, but you could pick a topic. Right? Or, wait, we could say which top questions and topics asked this month just to get an idea of what oh, fuck. Actually, I'm so I'm, like, not thinking about the fact that I need to ask about the intercom inbox. It's gonna just start talking about my own questions probably here once I know. I'm a little bit not thinking straight. It's Monday morning, so give me some leeway here. Monday. Exactly. In the intercom MCP. So I'm gonna say Intercom MCP. And then, basically, this is gonna go to my Intercom connection. Whatever client I'm connected to right now, basically, it's gonna pull up some top questions. So you can see here, it's like, okay. I'm searching the available tools for Intercom, and then it's looking now it's saying searching for the conversations this month. And now it's like, hey. Let me pull up recent conversations. And so this is what I was talking to Jess about as well. It's like, you could actually ask about the team member, for example, you could say, hey. What are the like, show me an me a sampling of closed conversations for this customer or for this team member this this month and also, you know, make a chart on, know, their stats or, like, CX score or whatever it is. And it should be able to pull that up for you. We can try in a in a second to do something like that as an example. This is this is great, Nathan. I'm I'm gonna. leave the stage so that someone else can chat as well as you, but I really. do appreciate your answer. I will be. back next. No problem. week. Thank you. Yeah. No problem. Yeah. Cool. I'd love to hear it when you come come back next week and share with us how it went. So Definitely. yeah. And if you need anything, Bye. reach out, like, in the community too. You can ask some questions in the actual forums in the community too. Cool. Appreciate it. Thank you. And thanks for coming on stage too. Okay. So I'm just gonna keep this going here just so we can see the results real quick. So, look, it pulled a 150 questions from March, out of five pages of results, and it's like, here's the top questions. Right? So this is just basically, here are the top topics right now, and it's pulling all those out. And this is, like this tool is crazy. Like, to me, like, if you haven't used the intercom MCP yet yet, you're really missing out on something that can really give you a lot of value. I show a lot of clients this, and they they go wild over it. And you guys you still need time to sit and work on it, but, like, here's here's some of the takeaways. Right? So now I can say, like, give me a list of 50 questions across these top topics, for example, that I can that I can use to test ZYN AI. And then it's gonna go and pull those questions for me. And then I can take those questions, and I can feed them in. Actually, I can I can ask it to make a CSV file of those questions, and you can import those if you go if you go back to one sec here? If we go back to Intercom I'll just go into my home here just to show you where that is. So so you go to Intercom and you go to FinAI, and then you go to the batch testing area, you'll see that there's a way to import these. So if you actually have a list of your top questions from some other process that you that somebody's done, then you can go into the test functionality here, and you can go upload CSV file. And you just drop your CSV file in there, and then you can create a batch of them as well. So you've got a few different ways to do this. If you've if you're just doing it the first time, you'll see this is this is the interface. But once you've done it the first time, then you'll see the other interface I was showing you. And then always, like, use this too. Like, see this give feedback button? A lot of areas of Intercom have this give feedback button. Highly recommend you use this to share feedback with the team because I don't know why it's not popping up. Maybe because I'm screen sharing at the same time. But, yeah, use this to share feedback with the team because it gets directly to the team working on this particular feature, and that is, like, super helpful. I've definitely had a lot of meetings with the batch test team, thanks to, like, just giving them more feedback, and, you know, we can improve this together. So I'm gonna jump into the comments to see questions anybody had. This is awesome. Thanks, Melissa. Is that app connection to Claude? Yes. Exactly. So good question. Let me show you, Terry. Yeah. No problem, Savannah. Appreciate it. The okay. So you were asking a really good question here. So how does this work? Right? So if you get the Claude desktop app, so claude.ai, for example feel like I should have, like, a referral code for Claude because I'm sending everybody over there. But go into go into your Claude desktop, go to settings, and then go to connectors. And then, actually, the cool part is Intercom has a built in connector with Claude, so you can just connect it. I have mine connected already. But the first time you connect it, basically well, I'll just disconnect it and show you how it works. So, basically, when you have it disconnected like this, you click connect, then it basically asks you to authenticate to Intercom the same way you normally do logging in to Intercom. So it'll go to the page, and then it'll say, okay. Choose your installation. So for me, I have a lot of a lot of client at Intercom, but you probably just have the one. So you log in, and when you authenticate that account, now that's the account that it's gonna use to communicate with. And, basically, you can see here it has access to people, companies, conversations, like, all that stuff that you normally would have. And if you use Claude CoWork, for example, versus using Claude Chat, you can actually have CoWork make files on your in your drive. So you can make a project related to your I think I have one related to this. It's related to the client. So I have a project related client. I could say, make a CSV file of x y z, and it'll put it'll save the CSV file on my computer for me so I can upload it. So you don't even have to, like, download it from the system. And I I often recommend using the OPUS 4.6 because it's the highest, you know, thinking model and then turn on the extended thinking. But sometimes if you're doing some, like, kinda menial stuff, stuff that's not as deep thinking, you can always use font Sonnet 4.6, which is a little bit more cost effective. I have the the max I have the clawed max plan. It's a $100 a month, but you can get the $20 a month plan. And, actually, the free plan does now support connections, so you can actually try for free before you, you know, upgrade. You can maybe run a couple reports to show your team so that they then can vouch for the idea of, like, paying for a Claude. But most of the clients that I work with have have upgraded it right away because they see how powerful it can be. You know? The other the other thing is if you if you have to use, like I don't know. If you're stuck using something like Enterprise or ChatGPT, we have had some people say that they were able to use ChatGPT and get a connector connected and use it that way too. But because Claude is the power the model powering Intercom as well, powering Fin, and that if there's already a connector here, it's just so much easier. So is the client data safe with this connector? That's a good question. I mean, Anthropic, Claude is the same company that is powering, like, the model that powers Fin. Right? So my my feeling is that, like, we know Intercom's already vetted this. Intercom provided the MCP. Intercom provided the the Intercom, MCP connection. I would say, like, in my opinion, it's as safe as you're gonna be with anything else you're using right now. What you can do if you wanna be a little bit safer is, like, in your prompts, specifically asking Finn not to include any per personal like, not include any, identifying data about the customers there. You know? So that would be one thing I would say. Yeah. Exactly. So, again, Steven, like like, whatever you're whatever you're thinking about doing, you can get, Claude Enterprise. I mean, sorry. Chat two p t Enterprise apparently has a connector, and some enterprise companies prefer using that. You can probably get an enterprise version of of Claude. But I would also say, like, just write your prompts in a way too that doesn't put your personal information over there. So, like, if you write your prompt and you can even ask AI to help you write a prompt. Like, hey. Help me write a prompt that doesn't bring my personal information over. Yeah. Yeah. ChatGPT does. One of our one of our previous, like, guests on the the office hours said that they were able to get it working. So yeah. I don't know if it's a out of the box connector, but you can connect it. You might just need the Intercom MCP data. I'll give you that link. There's a you might just have to tell it how to use the MCP. MCP one second here. Yeah. But I highly recommend this. The MCP is gonna change your life. If you do, like, if you do monthly reporting or quarterly reporting, like, you can you can draft up a much better report now. Like, intercom reports are very hard to navigate. I'm I'm just gonna be honest. Like, I'm not third party again. Like, I can speak openly about these things. Like, the reports are great if you like the reports, but if you're not a reports person, you don't know how to build it, and it's really kinda frustrating, just being able to type natural language and telling Claude, make a report about this. You know? Make a report about that. And then you can also say to Claude, hey. Give me five links to conversations that represent this stat, you know, this information. And then you can click on those individual links and get to them. So what I'd recommend is if you're a very, like, privacy conscious company, write your prompts in a way that say, you know, do not do not include any private information in the in the report. Linked to the conversation so that I can view the reports in Intercom. You know? Yes. You're not only are they difficult to interpret, but, like, when you actually create the report and you look at it and you don't you let you don't understand it, but you also feel like the data's wrong. And it's just because, like, the way Intercom was thinking about it and the way you're thinking about it is very different. Like, just like we were talking, Jess earlier was saying, like, we got a zero on that number. Why is that number zero? I don't know, and I don't have time to look into it and figure it out. Like, you got a whole you got a full time job to deal with. Like, you don't have time to troubleshoot this stuff. So just being able to go, boom. Like, Jessica, Jess. That's my favorite, the upside down happy face. Yeah. Just being able to go in and say, here's what I need. Show me a report for that. I mean, it's just so much better. I don't believe you. Exactly. Yeah. So, I mean, I think this is the way. I mean, I expect in the next maybe even maybe even something that we're gonna be telling us next week. There's a new there's a big, event next week to promote something new that's coming out. I don't know if it's part of this, but my feeling is that at some point, Fin should be integrated into Intercom in a way that we can ask these questions to Fin directly, and Fin can give us those answers, like like a Fin, you know, Fin directly for us. Like, if anybody uses Shopify, for example, if is anybody here right now using Shopify? They're a Shopify platform or Shopify e commerce company? If you do, there's, like, a little widget on the side of Shopify that you can sit and chat with about your data, and it's so great. Like, it's amazing. Like, you don't have to build a chart. You don't have to build a report. You just type the question in and get the answers back on the side. And I feel like the the Intercom's missing out with not having that particular feature for Finn. Like, just being able to say, yeah. Exactly. Like, Claude into the instance or just having, like, Finn, like, right there on the sidebar and say, hey, Finn. What what's our what's our, you know, what's our close rate today? Or, like, who's give me a report on this teammate or something like that. Like, there's no reason why we couldn't be doing that. But you can basically do it with Claude now. Right? So but, yeah, Jess, if you're using, if you're using Shopify and you haven't used the, their their little AI buddy I don't know what it's called now. I can't remember. But, yeah, it's it's you're you're missing out. I I can show you real quick because I think it's worth looking at because it's another example of how we can do the same kind of stuff. I do some work with this team called Omni. It's like a AI pendant, and they have where's my Shopify? I don't know if I can get to the Shopify. Let's see if I can. I think I'm I think I still have access to this. Let me see if I can real quick to show you something. I I use this with some clients, and I've showed them, and they're just, like they're tripping over it because it's like I hate building reports. All of the other the other tool that I would say is very similar. And don't worry, people. I'm gonna get to your questions any other questions in the chat, but I think this is relevant information to all of you. Sorry. I just gotta log in here real quick. Okay. So the let's see if this works. Never mind. Okay. I can't get in there. I'll show you a different one. So Posthog is another example of this. I don't know if anybody's using Posthog, but I use this on my on my on my product, and they have an AI. You if you use Zapier, similar types of things, like, Posthoc has an AI, and I can just go in here and ask a question about the data. Like, how many people signed up for the free trial or the, ah, man. The I can't type when people are watching me. Trial last week. And I go I can just go like this, and it's gonna go look at the data. Instead of me having to go find it, I can just type the question and look. I did the I did this yesterday, like, this week showing somebody. I said, like, make a rage rage click chart. It's like it's fun. Like, you can just go in and ask a question to your data and get an answer. Okay. So it says zero people signed up last week for the free trial. I don't know if that's just not finished building because I think that's not true. Maybe the wording I'm using is wrong. That's that's kind of interesting. That's interesting. Okay. So we'll go to the rage click one just for fun. So how many people rage click? Like, look at all the rage clicking over the last seven days. Like, that's a fun chart to look up. I'm actually kind of I'm gonna have to look at this later because I probably just worded this wrong. Like, sign ups. There's zero sign ups last week. Now I wanna ask them. I'm like, how many people are on the free trial? Let's see if it does it. Anyways, we don't have to worry about that. Just giving you an example of what it is. But Shopify has a little thing like this, post hoc AI similar, and you can ask those questions. And you can see the thought here. The user wants to know how many people are currently on the free trial, so it's going through and trying to find this. Okay. So it's trying to go through and figure it out doing doing a query. It still says free trial uses zero. I don't believe that that's true. So now I'm really like, this is not a great demo for you. But then, again, any report that is that way where you're like, at least you can see where it's looking and what it's doing, and I can figure out what's happening there. But just go back to your questions now instead of the quote instead of, like you know, demos never always work the same. Yeah. Let me see what other questions we got. And then if you have more questions, keep firing them off into the chat so I can, get there. So okay. I'm gonna answer Jessica's question because we kind of already answered it. Is cloud a better option than intercom reporting, or does it just have to give better summaries? I would say you can mix and match. Like, if it's an intercom report that is, like, a common report and you like it, they're good. You know? They're great. But if it's a if it's if you wanna get, like, some very meaningful data, I think using Clog is better. I give the example, which I'm building some tools for this right now, but I give the example. Like, as a start up, like, a a software as a service company, what's a better thing to look at? A a report that says, like, what's your first response time? Or or asking the question, like, what are the emotional jobs to be done in the inbox that we're not meeting? You know? If I can ask a question like that, like and, Jess, if you wanna come up on stage and talk about this, like, feel free to raise your hand. We can talk a little bit. But, like, for me, asking a question like, what are the emotional jobs to be done in the inbox, if you're not familiar with, like, the jobs to be done framework, that's gonna get me much higher ROI, improvements to support and customer service than just saying, like, how how fast is our response time or that kind of thing. You can only move that needle so far, but then, like, things like asking, you know, what's the what's the emotional job to be done gives you, like, all these really interesting things to dig into and and and work on and fix and stuff. So that's just my thinking. But I also think that if you're gonna build a report, the other thing I love about Claude is, like, you can ask Claude to, like, build out the charts. Like, it'll build out, like, visual charts and everything for you, and then you can say, put these in a Google Doc or put these in a, or make a PDF or put this into Notion. And if you use Notion, for example, you can ask Claude a whole bunch of questions, get a bunch of research, and then just say, a Notion page with this research that I can share with my team. And if you have the Notion connector connected, it'll make that page for you automatically. You don't have to like, you just take some of the work out of the thing. Okay, Gareth. You have a good one. I'm glad that you're, enjoying the session. Thanks for the feedback. Cool. I hope that helps answer some of your questions, Jess. I think we already kinda talked about this a little bit. And, yeah, you're welcome. Let me get on to, Steven. Steven said Steven. There we go. Steven asked, can Intercom make registering for these sessions easier? I will not remember to register weekly. Yeah. You can actually register, I think, for the event. Let me give you a better link. Intercom. I think it's just events.intercom. But one second. I think if you go to events.intercom and then you find our event, I think you can register for them all in one shot. Let me see here. Oh, no. I'm joining the event twice. One second. I wanna find the register link. Okay. Oh, you're right. This page. Okay. One second. I gotta find that one. The one that's in the community, I think, has, like, the register to multiple events at once. So this is a good time to actually mention this. I will say two things here about about about this. K. Let's see. Is this true? Just wanna make sure I'm saying the the right I'm saying the truth first before I give it to you. Yes. Okay. So this page here, you can find it in the intercom community. So let me come back to where you guys are. Okay. So there we go. Perfect. That's the one Marcel just posted. So if you go to two things I wanna do here. One, if you're not part of the Intercom community already, I'm gonna turn off the screen share, and I'm gonna say, see that meet the experts button right there? I'm gonna do my YouTuber impression. So go to this meet the experts button and then jump over to the community. And if you click on the intercom icon at the top of the community, then you can get to this page. Actually, why am I doing this? I can just screen share it. I can show you on the screen share. Okay. I'm gonna stop YouTube mode YouTuber mode and go here. Yeah. So click on this. Meet the expert. Oh, is it backwards? Is my video backwards? Dang. I didn't know that. Thanks for letting me know. No one's ever told me that before whenever I pointed the button, so I have to point the other way next time. Okay. So so when you click on meet the experts, you end up on this page. You can see me, Milan, and and Connor, Julian, Nur, and Joseph, like, all the experts. But you can also go to the intercom icon here, and that takes you back to the home page. And then you've got this community, meetup office hours link. And this one, although you have to jump, like, twice, which I don't understand, go there and then go here. And then when you end up on this page, which is probably the one Marcel posted to you, at the bottom here, you can register, but you can say select all, and it'll give you at least all the ones upcoming. I don't know if you get the ones after those several, but at least you get these you know, you don't have to register for each one of these ones individually. And, actually, you and I are on the same page because I told the Intercom team when they tell me about when they tell me about an upcoming event, why don't they just populate all these fields for me as well? Because they already know my name, my email address, the company. Like, why why do I have to fill that out every time? But that's the ongoing hassle that I continue to bring up. Again, you know, I'm not I I don't work for Intercom, so I can also tell you the truth. Like, I'm frustrated with some of those things too. You know? So so, Marcel, just to clarify, if I point over here, am I pointing at the meet the experts button now, or is it this way? It's over this way. Is that what you're saying? This way. Okay. Okay. Alright. Got it. I did not know that because on my screen, it's right here. So, anyways, it's good to know. I will point the other way next time. Okay. Let's see what other questions we have. Did that help, Steven? Okay. It's Claude. Okay. So okay. Patrice has a question here. Let me pull Patrice's question up. Patrice and if you wanna come up on stage and talk about this, feel free to raise your hand in the app, and we'll get you on stage here. I've only got, like, 15 less than fifteen minutes left, but I'm happy to stay beyond the top of the hour. So oh, you got options for IST. Okay. Cool. Oh, is that actually, is that not the time zone you're in? That's not good then. Is your is your workspace in that time zone, Steven Steven? Okay. So coming back to thanks for dealing with my ADE here. I'm all over the place, everybody, but I'm trying to answer all your questions. So Drew said, do you have any suggestions for streamline Fin optimization? For example, we have o [Transcript truncated]