How a Super Bowl support crisis led Kalshi to rebuild their stack on Fin | Community
Skip to main content

How a Super Bowl support crisis led Kalshi to rebuild their stack on Fin


Dianah Tripac
Intercom Team
Forum|alt.badge.img+3
Kalshi customer story: how a Super Bowl support crisis led to rebuilding support on Fin

 

✏️ New customer story: How a Super Bowl support crisis led Kalshi to rebuild their stack on Fin

 

Kalshi is the first federally regulated prediction market exchange in the US, and its support volume has grown from 30,000 tickets in 2024 to a projected 2 million in 2026. With a team of just six people running on email and macros, Operations Lead Shannon Magiera needed to scale without turning support into a 200-person call center. This is what happened when the biggest weekend of the year arrived just weeks after going live with Fin.

 

The headline numbers:

  • 80%+ of conversations resolved end to end by Fin
  • 80,000+ Fin resolutions every month
  • Median resolution time down from around 23 hours to a few minutes
  • Roughly 2 million conversations projected for 2026, handled by a small team plus Fin

 

The bits that make this one special:

  • Live in about 10 days. Kalshi works in a regulated space, so Legal and Compliance stress-tested every guardrail before launch, including rules on what Fin can say, what it must escalate, and how it handles financial questions without straying into advice.
  • Fin held up when the helpdesk underneath it didn't. On Super Bowl Sunday, 80,000 tickets arrived in 24 hours. Fin was already resolving 69% of the conversations it was involved in three days earlier, but the previous helpdesk hit its API rate limits and support went dark for 48 hours.
  • From crisis to a full rebuild. Kalshi moved its whole support operation to the Intercom Helpdesk in under three weeks, with Fin, Messenger, workflows and email routing all live.
  • Fin answers with real account context. For questions like "where is my money?", Fin checks the specific account, the market status and the deposit and withdrawal state, then answers with dates and amounts.
  • New roles, not fewer people. One teammate is now a full-time "Fin engineer", and the rest of the team spends more time on the conversations that need a human, including running trader happy hours with their users.

 

As Shannon puts it, "I get to eat dinner with my family again."

 

Read the full story: fin.ai/customers/kalshi

 

💬 Over to you: Have you stress-tested Fin ahead of a known traffic spike, or put guardrails in place for a regulated industry? Share what worked for you in the comments.