Skip to content

For anyone about to build an app

Find out if your app will sell. Before you build it.

Describe your idea. We build the landing page, the funnel and the ads, then put them in front of strangers who have never heard of you. In a few days you get one answer: build it or don’t.

No code. About $50 of ads. One-time licence at $149, never a subscription.

After 520 visitors, the answer

BUILD IT

3.46%

3.46% of visitors picked a plan and confirmed at your price. Above 2% means people will pay for this. Go build it.

Where 3.46% lands
You
Don't build
Promising
Build
0%1%2%5%+
520
Visitors
166
Started funnel
18
Said yes to paying
01

You write one paragraph

Describe the app the way you'd describe it to a friend. That's the whole input.

02

We build the test

A funnel carrying the price you'd actually charge, ads you can run today, and a tracked link for each one.

03

Strangers decide

We count who picked a plan and confirmed, then tell you in two words whether to build it.

What you get

This is what lands in your account.

Ten minutes after you describe the idea. Edit any of it before it goes live.

Your funnel

6 screens · written for your idea

Headline, walkthrough, device check, email step, paywall, confirmation. Every line editable, your real price on the paywall.

Night Owl

Stop scrolling. Start sleeping.

A twelve-minute wind-down for the nights your brain refuses to clock off.

Try it now →
Screen 1 · the hook
Night Owl

Your first seven nights are free.

12 months

$4.17

billed $49.99/yr

Monthly

$14.99

cancel anytime

Seven nights free, then $49.99/yr

Wind-downs that adapt to your sleep debt

Nothing to cancel by phone

Start free trial

Pre-registration. No card is charged.

Screen 5 · the decision
  1. 1Hook
  2. 2How it works
  3. 3Device check
  4. 4Email
  5. 5Paywall
  6. 6Confirmation

Your ads

3 – 5 angles

1 a.m. you vs
12-minutes-later you

Angle 02 · contrast

Primary text
You're not tired, you're wired. Night Owl walks your head down in twelve minutes. No breathing app voice, no whale sounds.
Headline
Asleep in 12 minutes
Description
Free for seven nights

Your links

one per angle

  • /p/night-owl?src=meta&ad=hook-01Instagram Reels
  • /p/night-owl?src=tiktok&ad=hook-03Spark ad
  • /p/night-owl?src=news&ad=sponsorNewsletter sponsorship

One tracked link per angle, and a link goes anywhere. You run the campaigns in your own ad account, so nothing here can spend your money.

  • Meta
  • TikTok
  • X
  • YouTube
  • Reddit
  • Newsletters
  • Podcasts

Try it

Drag the traffic. Watch the answer decide.

Pull the slider to send more visitors at a funnel. The answer starts at no verdict yet, then turns into a verdict once the data has actually decided. This is the engine your dashboard runs on, not a picture of it.

The verdict, as the engine computes it

Visitor → checkout

0.00%

NO VERDICT YET

0
Visitors
0
Funnel starts
0
Checkouts
<1%1–2%≥2%

95% range

Confidence

Reading

Nothing to report. The engine will not call anything under 30 funnel starts, and it has 0.

0 / 520Accumulating

It refuses to answer under 30 funnel starts, and it only reads high once the 95% range sits inside one band. If a segment beats the bar we name it instead of printing don’t build.

Where it works

One working audience can save the whole idea.

Every source, country, creative and audience is scored on its own, so an idea that only works in one place gets found instead of buried.

Checkout rate by segment

  • TikTok · United States · ad "screen-record hook"Works best here
    3.11%
  • Meta · United KingdomPromising
    1.47%
  • YouTube · United StatesNot viable
    0.25%
01%2%

0.85%

Blended · under the floor

3.11%

Best segment · over the bar

Promising

Verdict · not killed

And which screen loses them

One run, step by step

  1. 520Landed on your funnel

    Every confirmed human the ad sent. Bots never make it into the number.

  2. 166Kept reading past screen one

    32% kept. On par. A hero that holds a third of cold traffic is doing its job.

  3. 96Reached the paywall

    58% kept. On par. The walkthrough and the device check are not the problem.

  4. 18Reached for their card

    19% kept. Under par. A working paywall converts a quarter of the people who see it.

Read it in one line: the ads and the funnel are fine, the paywall is the leak. Fix the price or the promise before you touch the creative.

Price

$149 once. Then it is yours.

  1. $149

    Licences 1 – 500

    You pay this

  2. $199

    Licences 501 – 1,000

  3. $249

    Licences 1,001 +

The step happens when the licences sell, not on a timer. Every step buys the same thing: unlimited ideas, 50 tokens, every update we ship after.

  • Unlimited ideas. The licence is yours, not one project's.
  • 50 tokens, and they don't expire
  • Every update we ship afterwards

30-day money-back guarantee. Tokens you already spent don’t count against the refund.

Objections

The four questions everyone asks first.

Isn't a paywall for an app that doesn't exist dishonest?

It would be if anyone were charged. Nobody is, and no card is ever collected. The checkout step asks for a name and an email so you can tell them when it ships, and it says on the screen that they are pre-registering for an app that is still being built. What you keep is their decision at your price, and that is the whole point.

How much ad spend does a verdict take?

A first verdict unlocks at 30 funnel starts, usually around 100 visitors or $50 to $100 of traffic. High statistical confidence costs more than that. The dashboard tells you roughly how much more, instead of letting you assume you’re done.

Do you launch the ads for me?

No, and we won’t pretend otherwise. You get the creative, the copy and a tracked link per angle, then launch in whichever ad manager you already use. The upside: those links work on TikTok, X, YouTube, Reddit, newsletters and podcast reads exactly as well as they work on Meta.

What if the verdict says don't build?

Then it cost you a weekend and a hundred dollars instead of four months. You keep the licence, the tokens you haven’t spent, and everything the run taught you about the angle. The next idea starts on Monday.

All twelve answers

Four months of building, or one week of finding out.

Start a test

30-day money-back guarantee