What Eden taught me about first contact
Running our own web chat before selling one to anybody else. The failure modes are not the ones the demos prepare you for.
Eden answered chat on this site, and Hermes still handles enquiry intake and follow-up. Both exist because I needed them before anyone else did, which makes them the only two things here that have met real traffic. Eden is off the front door at the moment — the lessons below are why.
Building for yourself first is not humility. It is the cheapest way to find out what you do not know, on a customer who will forgive you.
The failure modes nobody demos
- Confidently wrong about your own business. The first serious error was not a hallucinated fact about the world. It was a confident, plausible, incorrect statement about what SUNVEX does. Every demo tests general knowledge. Nobody tests whether it knows you.
- Helpful when it should stop. Someone asked something I should have answered personally. It answered. Well, even. That was the wrong outcome, and nothing in the transcript looked like a failure.
- Cost is spiky, not linear. One person having a long, curious evening costs more than a great many short sessions. You do not discover the shape of that from a pricing page.
- Silence reads as broken. A couple of seconds of nothing and people leave. Latency is not a technical metric here, it is the whole impression of competence.
What changed as a result
Scope got narrower and the handover got better. Eden answered what was safe and said so plainly when something needed a person — not as an apology, as a routing decision. The rate limiting exists because I watched what an unbounded conversation costs. Whether a chat box is the right front door at all is the question that took it down for now.
The interesting engineering is not in what it says. It is in knowing when to stop talking.
Why this matters for what comes next
V-001 is the same problem wearing different clothes: work that reaches a customer on behalf of a business that cannot afford to get it wrong. Everything above was learned on my own front door, at my own expense, before it becomes a promise to somebody else.
That is the whole argument for building for yourself first. Not that it proves the product works. That it means the first person a failure costs is you.