Intercom uses Apify to ingest data for customer chatbot Fin

Intercom’s flagship customer service bot Fin was initially limited to knowledge bases hosted by Intercom Articles. To answer customer questions accurately, Fin needed access to external customer data, including documentation and blog posts. Building an in-house web crawler would have distracted from core product development.

Apify delivered a production-ready cloud-based web crawler that was accessible via API in one month. The solution automatically crawls customer websites, handles JavaScript-rendered pages, bypasses anti-scraping measures, and filters out irrelevant content like navigation and ads. Now, Fin can be fully trained on customer content in under 15 minutes.

Fin AI Agent, Intercom engineering team, Intercom customers’ support teams

1

Website Content Crawler, browser fingerprinting, proxy infrastructure, JavaScript rendering

Head of functional/operational unit

Full implementation

Joint management between Apify and Intercom

Engineering team with API integration knowledge

API access to Apify platform

Fin can now access and understand customer documentation and blog posts, providing accurate answers based on customer-specific content. The solution enables rapid deployment, with Fin being live and fully trained in under 15 minutes.

**Quantitative Results**: – 24 million+ conversations successfully closed by Fin – 73% resolution rate – Under 15 minutes to fully train on customer content – 99.95% platform uptime

**Customer Feedback**: “Apify gives us rapid access to our customers’ public content. Our Fin AI Agent can be live and fully trained on customer content in under 15 minutes.” – Gearoid Miggan, Engineering Manager for Knowledge “Apify has been an awesome launch partner. Their technology and team helped us extend Fin with external knowledge from customers’ websites, and ultimately bring our new AI-powered chatbot to the market faster.” – Brian McDonnell, Director of Engineering

Rapid delivery (one month pilot), complete solution including anti-scraping bypasses, automatic content filtering, API-based integration enabling focus on core product development.

Handling dynamic JavaScript-rendered content and bypassing anti-scraping measures while maintaining high performance and reliability.

Consider specialized, external providers for web crawling infrastructure rather than building in-house. This will allow engineering teams to focus on core product features while getting ready for production faster.