Why we decided to discontinue Stemma

A few months ago, Robert and I started building Stemma: an AI voice assistant designed to answer phone calls for businesses. The initial idea was simple: many small businesses miss calls because they’re busy doing the work they were hired to do. What if AI could answer those calls, handle basic conversations, and convert them into bookings?

Over the course of the project, we built a prototype, pitched the idea, spoke with potential customers, developed pricing strategies, discussed founder agreements, and explored what it would take to turn Stemma into a real company. We even had people reach out to us asking about pricing, which was both exciting and validating.

So why stop now?

The short answer is that Robert and I reached the same conclusion: while we believe there is a real opportunity in AI voice agents, we no longer believe that Stemma is the opportunity we want to dedicate the next years of our lives to pursuing.

One of the biggest lessons I’ve taken away from this experience is that there is a difference between asking:

“Can we build this?”

and asking:

“Do we want to build this?”

The first question is technical. The second is personal.

Over time, I realized that although I enjoyed building the product, experimenting with voice technology, and thinking strategically about the business, I was much less excited by the prospect of spending years selling into this particular market and solving this specific problem.

That doesn’t mean Stemma was a bad idea.

In fact, one of the surprising lessons was that a startup can show encouraging signs and still not be the right fit. We had customer conversations, a working prototype, and indications that there was genuine interest. If anything, the experience made me appreciate how important founder-market fit is alongside product-market fit.

I don’t view Stemma as a failure.

It taught me more about startups in a few months than I could have learned from reading books alone. I got hands-on experience with customer discovery, pricing, pitching, rapid prototyping, working with a co-founder, and the uncertainty that comes with trying to build something from scratch.

Most importantly, it helped me understand a little more about what energizes me. I enjoy building AI products, shipping quickly, and solving technical problems. As I continue my journey in AI and entrepreneurship, those are lessons I plan to carry with me.

I’m grateful to everyone who gave feedback, challenged our assumptions, showed interest, and supported us along the way.

The world keeps spinning.

Stemma may be ending, but I don’t think this is the end of my entrepreneurial journey. If anything, it’s probably just the beginning.