Fanatics
Multi-agent customer support for Fanatics Betting and Gaming
Fanatics reported approximately 56% better support containment after deploying AI agents
Problem
Fanatics Betting and Gaming faced growing support demand and different betting rules across states. Sensitive gambling-related conversations also required human escalation.
Approach
Its engineers built a multi-agent system on AWS. A Claude supervisor accessed support knowledge and account tools, while a separate classifier screened for responsible-gaming concerns. High-severity concerns triggered transfer to a person with the conversation history.
Results
Takeaway
Human escalation remained part of the automated workflow. The reported improvements were relative changes from internal measurements, rather than absolute containment levels.
Categories
- Size
- Large enterprise
- Industry
- Media and entertainment
- Business function
- Customer service
- Innovation type
- Process innovation
- AI technology
- Generative AIAI agents
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
AWS described the deployed support architecture. It reported approximately 56 percent better containment and approximately 53 percent better resolution rates, citing Fanatics' internal measurements.
Full details are in the original sources.