First-line support answered around the clock

Intercom's Fin agent answers first-line support from your help centre, carries out routine actions such as refunds and account changes within your policies, and leaves the complex cases to your team

The problem

A SaaS or online retail team in Singapore or Bangkok with two or three support staff gets the same questions after hours and across time zones: where is my order, how do I change my plan, can I get a refund. Each one waits for the morning shift, and the people who could fix the hard problems spend their day on the easy ones.

How the agent works

1. It starts

   The team turns on Fin inside its existing Intercom workspace and points it at the knowledge it already has: the help centre, internal articles and past conversations.

2. The agent works

   Fin draws on that knowledge base to answer customer messages in the messenger around the clock, in the tone of voice and answer length the team sets, and it follows the policies and procedures the team writes down, for example when a refund is allowed. Where the team connects an action, Fin can carry it out, such as processing a refund or changing an account detail, instead of only explaining how. Anything outside its knowledge or its policies goes to the human team, who handle the complex cases. The case study describes this split of work, not the handoff mechanics, so the team tests the handoff rules on a copy of real conversations before switching on.

3. The result

   Fin's analytics show which questions it resolved and which it passed on, and the team uses that list to write the next help articles.

Worked example

Anthropic's case study on Intercom describes Fin, built on Claude, as answering customers instantly and 24/7, drawing on the company's entire knowledge base, adapting to each business's style and policies, and taking actions such as processing refunds and managing account changes. Intercom's figures in the study: a 51% average resolution rate out of the box, resolving up to 86% of a customer's support volume, support in over 45 languages, and over 25,000 customers. Named examples include Synthesia with over 6,000 conversations resolved and over 1,300 hours saved within six months, and Fundrise automating over 50% of total support volume within three months with a 95% response accuracy rate. These are Intercom's and its customers' numbers; your own resolution rate depends on how complete your help centre is.

Sources

• https://claude.com/customers/intercom