How Hatz AI uses Fin to deliver its best customer experience
Hatz AI helps small businesses adopt and manage AI. Selling through the IT providers those businesses already rely on, Hatz AI gives these managed service providers a single, governed platform to package, sell, and run AI tools for their own clients. It’s a product with a huge range of models, settings, and controls, serving a fast-growing user base, many of them integrating AI for the first time.
Growth came fast. A complex, fast-changing product and a wave of first-time AI users across more than a thousand partners and tens of thousands of end users sent support volume climbing faster than the early setup could absorb. “We went from knowing all our customers by name to ‘who is this?’ almost overnight,” says Jimmy Hatzell, founder and CEO of Hatz AI. “It was just a shared inbox, and it got totally unmanageable.”
Jimmy had a clear conviction from the start: an AI company should be able to use AI to help with customer support. “We’re an AI company, and we believe that AI has the ability to address a company’s specific needs and provide the right support if you build it right,” he says. Hatz could have built its own support agent, but chose Fin, an AI customer agent, instead. The decision came down to speed and reliability. When Jimmy needed to get up and running over a single weekend, Fin made that possible, with reliable infrastructure he didn’t have to maintain himself. Fin’s Early Stage Program was the way in.
Hatz deployed Fin alongside its natively integrated AI-powered helpdesk, Intercom, and put Fin in front of every customer. The bar Jimmy set was simple: give people a complete, accurate answer immediately, and when something needs a human, hand it over fast. Fin met it from the start, resolving the bulk of Hatz’s conversations early on and quickly becoming the team’s first port of call. That includes Jimmy, who now asks Fin about product edge cases he once had to dig through the codebase to answer.
Fin is involved in around 90% of conversations and resolves around 80% of the conversations it touches without a human stepping in, even when customers ask complex product questions. Their CX Score, an AI-generated measure of answer quality across every conversation, peaks on the conversations Fin resolves end to end, at around 80%. It reflects what customers value most: accurate and tailored answers specific to their questions, without waiting or being passed around.
“Being able to pull in real user conversations with Fin and use them as the data point for what we build has been my most important feature.”That quality has held as Hatz scaled. Support conversations roughly doubled in four months, from around 600 to 1,300 a month, while the human team held steady. Fin has become more than support: by surfacing what customers actually ask about, it now feeds Jimmy’s product roadmap. “Being able to pull in real user conversations with Fin and use them as the data point for what we build has been my most important feature,” he says.
For Jimmy, this is just the start. “My approach to support is ‘AI maximalism’. I want people to interact with us and think: wow, I didn’t know AI could do this well.” His advice to other founders is refreshingly simple: “Just get started. Get your knowledge base right, and set it up.” For Jimmy, the ambition goes beyond automating support. He wants AI to become part of how Hatz operates as a company, while giving customers a firsthand experience of just how well it can work.
