I built an app - now what?

Nothing is wrong with you and probably nothing is wrong with the product. You did the hard part and then hit a wall nobody mentioned: building and being found are two separate projects, and only one of them got easier.

The good news is that the next month is genuinely sequential - there's a right order, and following it badly beats following a better plan out of order. The bad news is that none of it works in a day, and the first three weeks produce numbers that look like failure if nobody tells you what normal is. So here's normal, and here's the order.

This week has no creativity in it at all. It is a checklist, it takes an afternoon or two, and skipping it makes every following week unreadable.

• Check that machines can read your site. View source and look for a sentence you can see on screen; load the page with JavaScript off. If your content isn't in the HTML you're invisible from birth, and nothing else here matters yet. Full walkthrough: is my site invisible to Google?
• Write the one sentence. What it is, who it's for, what it replaces. Plain nouns. "A time tracker for freelance designers who currently invoice from a spreadsheet" beats anything with the word "seamless" in it. Put it above the fold, in text, not in an image.
• Fix titles and descriptions. Every page gets its own, describing that page. These are your search results - the only copy some people will ever read.
• Install measurement. Analytics, Search Console verified, sitemap submitted, and one activation event defined: the specific moment a user gets the value you promised. Not signups. The moment it works for them.
• Cut to one action. Generated pages arrive with four equal buttons. Pick the one you actually want and repeat it. Four choices is the same as none.

That's the week. If you finish early, don't start marketing - go read the receipts so you know what the next three weeks are supposed to look like before you live through them.

The week everyone skips, because it's uncomfortable and doesn't feel like leverage. It is the highest-information week you will have all month.

Find ten people who have the problem you built for - in the forums, subreddits and Discords where they already complain about it - and start conversations that aren't about your product. Answer things. Be useful. Where it's genuinely relevant, mention what you made. Ten specific people beat a thousand strangers, because ten people will tell you why they didn't sign up and a thousand strangers just leave.

At the same time, message anyone who has already used it and ask two questions. What were you doing before this? What nearly stopped you signing up? The answers are usually a sentence you should have written on your homepage, in their words rather than yours.

One rule matters more than the rest of this section: do not post your link cold. Communities remove it, and removal is the good outcome - the bad one is being permanently the person who did that. Earning the right to mention something takes weeks of being present, which is why it starts in week two rather than week ten.

Two. One slow and compounding, one fast and answering. Each with a fixed window and a number attached before you begin.

The slow one is almost always search-shaped: one page that answers a question your buyer genuinely types, with the answer at the top instead of five paragraphs of preamble. It takes months to pay and then keeps paying. In 2026 that also means AI answers, since a growing share of buyers ask a model and never see a results page.

The fast one is direct: outreach, communities, partnerships, wherever your buyer already gathers. It doesn't scale and that's fine - its job is to tell you this week whether anyone wants this, while the slow channel warms up.

Write the window down before you start: what you'll do weekly, for how many weeks, and what would count as a signal. Most channels get abandoned around week three - before any of them could have produced an answer - then retried six months later from zero. A channel you quit at week three is a channel you paid for and never read the result of. The channel picker decides it for you in a minute.

Now the measurement you installed in week one earns its keep. You're looking for which of three things is true, not for good news.

For calibration: across four sites launched in 2026, first-weeks impressions ran between roughly 2,200 and 16,500, with click-through rates of 0.1% to 0.6% at average positions between 12 and 49. Those are real numbers from real Search Console exports, published in full on the receipts page including the unflattering ones. If your month looks like that, your month is fine. The most expensive mistake available in week four is concluding the site failed, rewriting everything, and resetting the clock.

Said plainly, because these are the four weeks most people spend on the things below instead.

• "Build it and they will come." Dead, and the arithmetic is why. One AI build platform reports on the order of a million new projects a week, roughly 63% from people who've never written code; the App Store added around 560,000 new apps in the first half of 2026. The number of slots on a results page didn't change. Being good is the entry fee now, not the strategy.
• A launch platform as the plan. Worth doing once, worth no anxiety - one day of contested attention that doesn't compound, and a single point of failure you don't control.
• Dropping your link in communities. Removed at best, remembered at worst. The slower version works reliably: be useful there for weeks first.
• Redesigning the landing page again. It's the work you already know how to do, which is exactly why it's the trap. Without measurement, version four isn't better than version three - it's just newer.
• Ads before the page converts. Paying to send strangers to a page nobody understands buys proof you already had for free.
• Waiting for something to go viral. Not a plan. A lottery ticket with a deadline attached.

You built a working product. Whatever you used to do it, that happened, and most people who intend to ship something never do. What you're missing isn't intelligence or effort - nobody ever taught you the other half, and the tools that made building nearly free have no reason to mention that distribution got harder at the same moment.

The other half is learnable in about a month, and it's mostly unglamorous: be readable by machines, say what you are, talk to people with the problem, pick two channels, measure one honest number, don't quit in week three. To find your actual bottleneck first, the two-minute diagnostic names it, and the full post-build guide covers each gap in order.

The 30-day post-build playbook goes week by week with the specific checks, the numbers to watch, and what to do when a week doesn't go to plan. One email, no drip sequence, no course.

Or get the 30-day post-build playbook Send it