Customer support · Production

Clay scales customer support to 90% AI involvement with Intercom Fin AI Agent

The problem

Clay's three-person support team managed all customer queries manually through a Slack community with no automation, causing customers to wait four to six hours for answers, leaving agents stressed, and making it impossible to scale support as the customer base rapidly grew.

Workflow diagram · grounded in source
1
Customer submits support query
trigger
“Fin is enabling Clay to support customers on multiple channels”
2
Fin AI Agent engages customer
ai_action
“Fin is now involved in 90% of Clay's conversations across all channels”
3
Fin Tasks connect internal systems
integration
“With Fin Tasks and data connectors, we were able to configure it to answer tier two and even tier three questions, and connect it with our internal systems for managing things like refunds. That's something we haven't been able to automa…”
4
Fin resolves or routes to human
routing
“resolving up to 50% of them. With Fin's help, Clay's support team has been freed up to focus on more complex queries”
5
Human agents handle complex queries
human_review
“Our team is now answering the most complex, strategic, technical questions and agents aren't being overloaded with tickets”
6
Performance monitoring and feedback
feedback_loop
“To measure the success of Fin, we track a lot of the normal things like resolution and involvement rate”
Reported outcome

Fin AI Agent is now involved in 90% of Clay's support conversations and resolves 45–50% of them autonomously, freeing the support team to focus on complex queries while customers receive answers almost instantly.

Reported metrics
Fin involvement rate90%
Fin resolution rate (current)45-50%
Fin initial resolution rate20%
Monthly support query volume7,000 per month
Show all 7 reported metrics
Fin involvement rate90%
Fin resolution rate (current)45-50%
Fin initial resolution rate20%
Monthly support query volume7,000 per month
Pre-Fin customer wait timefour, five, even six hours
Post-Fin customer response timeno longer need to wait even 10 seconds
Counterfactual headcount needed without AIdouble the size of the team
Reported stack
Fin AI AgentIntercomSlack
Source
https://www.intercom.com/customers/clay
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Fin AI Agent is now involved in 90% of Clay's support conversations and resolves 45–50% of them autonomously, freeing the support team to focus on complex queries while customers receive answers almost instantly.

What tools did this team use?

Fin AI Agent, Intercom, Slack.

What results were reported?

Fin involvement rate: 90%; Fin resolution rate (current): 45-50%; Fin initial resolution rate: 20%; Monthly support query volume: 7,000 per month (source-reported, not independently verified).

How is this customer support AI workflow structured?

Customer submits support query → Fin AI Agent engages customer → Fin Tasks connect internal systems → Fin resolves or routes to human → Human agents handle complex queries → Performance monitoring and feedback.

Related customer support cases