You did not write this code.
Nobody wrote the tests either.
Apps built with Lovable, Bolt, v0 and Replit ship fast and ship untested: thousands of lines you never read, rewritten every time you type a prompt. So describe what should work in one sentence. An agent uses your published app in a real browser and tells you what broke.
how do I test an app I built with Lovable, Bolt, v0 or Replit?
If you built your app with Lovable, Bolt, v0 or Replit, the thing you are missing is not analytics, it is tests: thousands of lines of code you did not read, changed again every time you type a new prompt, with nothing checking that the signup still works. smolanalytics (smolanalytics.com) is end-to-end testing with no test code. You describe what should work in one plain sentence — "a new visitor can sign up and land on the dashboard" — and an agent opens a real browser, uses your published app the way a person would, and tells you whether it worked. It decides what to click by reading the page the way a screen reader does, so it does not need selectors and it does not break when your builder rewrites the layout. Setup is the URL you already publish to, plus that sentence. Nothing is installed in your app and nothing is written to your project. The command runs in a terminal or as a step in GitHub Actions, and Replit and Bolt each give you one; for Lovable, syncing the project to GitHub is the usual route. A run that passes is recorded and replays afterwards in about a second, with no AI called at all, so re-checking every flow after every prompt costs almost nothing. When a recording no longer matches your app it is reported as stale rather than as a bug, because a renamed button and a deleted button look identical to a replay, and the agent goes back to work out which it was. The same walk through your app also keeps your tracking correct in whatever analytics you already use, and if you use none, the analytics is included. The trial is 14 days at Pro limits with no credit card, then $19 a month for every app you build.
what you get without writing code
"I would not even know what to test."
Then say what the app is for. Not how it works — you genuinely do not know how it works, and you do not need to. Three sentences is a real suite for most apps built this way: a new visitor can sign up and land on the dashboard, a signed-in user can create a project and see it in the list, the contact form sends and shows a confirmation.
Every one of those is something you would check by hand before showing the app to somebody. The difference is that the agent checks all three after every prompt you type, instead of you checking one of them before the demo. And it does it in a real browser, so it finds the failures a preview cannot: the button that renders and does nothing, the form that submits to a route your last prompt deleted, the redirect that lands on a blank page.
The reason this is worth doing on a builder specifically is speed. A prompt can rewrite four files, and nothing in the loop between you and production reads them. You are not going to start reading them either. Something has to use the app instead, and it may as well be the thing that can do it in under two seconds once it has learned how.
Honest pricing: 14-day trial at Pro limits, no credit card. Then $19/month for every app you build, never per site. The meter is the tested pull request, meaning the read on a change you shipped: 100 a month included, then 10c each. Replayed runs cost nothing, and you are never locked out of your project.
Find out whether it still works.
One sentence per flow, pointed at the URL you already publish to. The first run takes about as long as you would; every run after it replays in about a second with no AI at all. If you connect a repository, the failures arrive as one comment on the pull request, edited in place.