Three emails a solo founder sends, and what came back
AppX now sends three short emails from me: when your first app is ready, when you built one and did not come back, and when you signed up and never started. 56 sent so far. Here is what they say and the mistake that means I cannot tell what came back.
Ilyos Olimov · · 3 min read
I read every signup. For a long time that was the whole CRM: a database query in the morning and, if someone looked interesting, an email written by hand.
That stops scaling at about ten signups a day, which is where an ad test put me last week. So I wrote the three emails I kept writing by hand, and let the server send them. They are short, they are from me, and they are the same for everyone. Here they are, with the numbers.
1. "Your app is ready"
Sent within a day of someone's first successful build. It says, roughly: your first app built, here is the link, the next step is to open it on your phone and change one thing by chat. One paragraph, one link, my name.
Sent so far: 21.
The point of this one is not information, the user already saw the app. The point is that a person noticed. Most products send a welcome email when you sign up. Nobody sends one when you actually did the thing.
2. "Your app is still where you left it"
Sent once, after three days of silence, to someone who built an app and did not come back. It names the app. It says the app is still there, nothing expired, and asks one question: what stopped you?
Sent so far: 11.
I expected this to be the awkward one. It is actually the most useful, because the honest answer to "what stopped you" is the roadmap.
3. "Your 100 AppX credits are waiting"
Sent one day after signup to people who created an account and never started a project. It says: you made an account yesterday, you have not built anything, your 100 free credits are still there, describe your app in one sentence in English, Russian or Uzbek, a first build costs about 30.
Sent so far: 24, all from the ad-test cohort. I limited it to accounts created after October 15 on purpose. Someone who signed up in July and left does not need a nudge in October.
The mistake
Every one of those 56 emails went out from a no-reply address.
The templates have a line, "reply to this email and I will read it", that only renders when a reply-to address is configured. I never configured it. So the line was silently dropped, the emails ended with my name and no way to answer, and any reply that someone did send went into a mailbox nobody reads.
Which means the honest answer to "what came back" is: I do not know. The emails were sent correctly, the claims in the database say so, and the deliveries did not bounce. But the question I most wanted answered, "what stopped you", had nowhere to go.
I fixed it on October 21. The reply-to address is now my own inbox, and the "reply to this email" line renders again. Everything sent from here on can be answered. I will write a follow-up with what people actually said, once there is enough of it to be honest about.
What I would tell another solo founder
- Write the three emails you already write by hand. Those are the right ones.
- Send them from a person, in plain text, one paragraph.
- Check the reply path with a real reply before the first send. I did everything else right and missed the only part that gives you information back.
- Keep the counts honest. 56 emails is not a funnel. It is a start.
Keep going
More practical ways to move your app forward.
- How to describe your app to AI: a prompt template
Write a clear AI app-builder prompt with a copyable template, a practical example, acceptance checks, and focused follow-up edits.
- How to turn a client brief into a mobile app prototype
Turn a client brief into a mobile app prototype with a copyable AppX prompt, a practical test checklist, and a focused client-feedback workflow.
- The 5-backtick fence: stopping prompt injection via Markdown
We ship user input into an LLM architect prompt every turn. A reviewer caught that a 4-backtick string from the user could close our 3-backtick fence and run arbitrary instructions on the model. The fix is one weird Markdown trick.
- Investigators before builders: a 2-phase LLM workflow
Most teams dispatch coding agents straight to the build. We run a read-only investigator phase first — and the findings have rewritten our plan in 30-40% of cases. Here's the shape, the brief format, and what to do when findings contradict the plan.
- Step 0: delete dead code before you refactor with an LLM
Every byte you load into an LLM agent's context is a byte it has to keep track of. Dead code looks like signal — and agents preserve, extend, or mimic it. Our rule: a pure-deletion commit before any structural refactor of a file over 300 LOC.
Ready to try your idea?
Start with a clear description. AppX will help you turn it into an app you can test.