Kartik Kant / All lessons Chapter 5 · Finding your first users
Lesson 5.5

Keeping it going

The apps that work out usually aren't the ones built by the most talented person. They're the ones where somebody kept showing up after the interesting part finished.

That's most of what this last lesson is about, because the failure mode for solo builders isn't running out of ability. It's running out of steam, or spending months on the wrong thing while feeling productive.

The pull toward the comfortable work

The most useful thing I can tell you, and it's the thing I got wrong.

Building is comfortable. It's visible, you control it, progress is obvious, and nobody can reject it. At the end of a day of building you can see what you did.

Selling, launching, talking to strangers, asking for money are all uncomfortable. They involve other people, and other people can say no.

So there's a constant quiet pull toward building, and it feels like productivity because you are genuinely working. You can spend six months busy and end up with a polished app nobody has seen.

The check that helps: ask what's actually stopping this from succeeding right now. Usually the honest answer isn't a missing feature. If your app works and nobody knows about it, another feature is avoidance wearing a helpful face.

Loops beat pushes

Early on, every user comes from you doing something. You post, you message people, you show up somewhere. Stop and it stops.

The things that compound are the ones where using the app produces more users, or where work you did once keeps working. Someone shares their result and their friend signs up. An article you wrote a year ago still brings people in every week.

Those take longer to start and don't stop when you do. Worth building one deliberately once you know people actually stay, and there's no point building one before that, because a loop that brings people into an app they'll abandon just speeds up the leaking.

A rhythm that survives bad weeks

Motivation is unreliable and comes and goes for everyone. Habits carry you through the weeks it doesn't show up.

Something roughly like this works for a lot of solo builders.

Each day, one uncomfortable thing before the comfortable ones. Message someone, post something, follow up. It takes twenty minutes and it's the part that would otherwise never happen.

Each week, look at your handful of numbers, pick the single most important thing, and talk to at least one person who uses the app.

Each month, step back and ask whether the thing you're working on is still the thing that matters. Small enough to keep doing, structured enough that the uncomfortable work happens whether or not you feel like it.

Knowing what to do next

You'll have more ideas than time, permanently, and that doesn't improve.

The question worth asking is what's most limiting things right now. Not what's most interesting or most requested.

If people arrive and don't stay, features won't help. If people stay and there aren't enough of them, more features won't help either. If nobody's paying, another free feature definitely won't help.

Most of the time the constraint is obvious once you ask directly, and it's usually not the thing you were about to build.

When to stop

Worth saying, since most advice pretends quitting is never right.

Some ideas don't work. Not because you executed badly, but because the problem wasn't painful enough or the people won't pay. Continuing out of stubbornness isn't a virtue, it's just expensive.

Signs it might be time: you've genuinely tried to reach people and they don't stick, the ones who do won't pay, and you've stopped learning anything new from continuing.

That's not failure in a meaningful sense. You now know how to build and ship something, which you didn't before, and the second one takes a fraction of the time. Almost nobody's first attempt is the one that works.

The pace

Software takes longer than it looks from outside. The overnight successes usually had two years you didn't see.

Which matters practically, because burning out in month three ends everything. You are the whole company right now. A pace you can hold for a year beats a sprint that stops in April.

A quick story. I want to end with the honest version rather than something encouraging.

I built a complete app. Ninety days of content, custom artwork, a backend that works offline and syncs, onboarding I rebuilt more than once, payments, a website, a privacy policy. It's genuinely good work.

It hasn't launched. It sat at submittable for a long stretch while I improved things that were already fine.

Every piece of advice in this course is something I either learned properly or paid for by getting it wrong. This last one is the expensive one. Building is the part I'm good at and the part that feels safe, and it will absorb infinite time if you let it. The uncomfortable work is where the outcome actually lives.

If you take one thing from all of this: ship it earlier than feels reasonable. You'll learn more in a week of it being real than in three months of it being nearly ready.

If you want help staying on track

Work this chapter with Claudea prompt to run on your own project
I'm building an app on my own and I want to make sure I'm working on the right things. Please be direct with me rather than encouraging. Where I am: [describe honestly, including whether you've launched] What I'm working on now: [describe it] What I know about my users: [numbers, or "none yet"] Please: 1. Tell me what's most likely limiting my progress right now, based on what I've said. 2. Tell me honestly whether what I'm working on addresses that, or whether it's comfortable work that avoids something harder. 3. Suggest a simple weekly rhythm for someone in my position, including the uncomfortable things I'd otherwise skip. 4. Tell me what evidence would suggest this idea isn't working and I should move on.
Open Claude ↗

Point two is the one to sit with. Ask it periodically and answer honestly, and you'll save yourself months.


That's the whole course. You've gone from an idea, through building, charging, shipping and finding users, to keeping it alive.

The only part nobody can do for you is putting it in front of a real person. That one's yours, and it's earlier than you think.

Go deeper: Chapter 51, Growth Loops and Experimentation and Chapter 53, The Founder's Operating System

Back to the start of the course →

Useful? Share this lesson