ShouldShip

$19/mo → 3 weekly “build this” picks

Not traffic. Pre-screened starts.

Sample issue

Three starts. This density.

Each pick, one screen: Job → Why → 72h → Evidence → Caveat.

Pick 01 of 03

Stripe → signed ZIP for solo template sellers — pin “your Stripe + one ZIP”.

Why

Marketplaces already take ~10%+ to host the file.

The thin job is checkout on your Stripe and one signed ZIP.

72h

One SKU, Checkout session, webhook emails a 24h signed URL.

10 template sellers. If fewer than 3 pay, stop.

Caveat
Tax / MoR is why people stay on Gumroad — do not sell VAT.

Pick 02 of 03

Fail-closed per-tenant AI credit hard-stop — refuse the next call, <1h install.

Why

Spend limits are org/project, not per customer.

Meters report after the call. The job is a hard refuse before the next token is billed.

72h

Middleware returns 402 when the tenant balance hits 0.

Install in under an hour. If refuse is not the default path, stop.

Caveat
Refuse, not a meter — invoicing or a billing suite breaks this pick.

Pick 03 of 03

Failed-payment recovery inbox for <$10k MRR Stripe SaaS — vs Churnkey ~$250+.

Why

Churnkey starts at $250/mo and bundles cancel flows.

Under $10k MRR the job is a founder inbox of this week’s failed invoices.

72h

Webhook → digest of past-due invoices with recovery links.

Run it on your own Stripe. If you will not write this week’s failed charges, stop.

Caveat
Stripe already retries; thin inbox, not a Churnkey clone.

Who pays

English coding indie. Wrong-build weeks already spent.

You can ship. Nights go to HN, Reddit, and ChatGPT. The payment is for 3 pre-screened start candidates — not another tab.

  • Can ship code

    Solo English-market indie. The stall is what to build next.

  • Browsing is the default

    HN / Reddit / ChatGPT already exist. This is not those tabs priced at $19.

  • Weeks already burned

    The exchange is 3 starts with evidence. Not a longer list.

What $19 buys

3 pre-screened start candidates. Evidence. A 72h path.

ChatGPT emits a list. A pick is a job, why it is worth starting, links you can open, and a 72-hour start path. Not a feed.

  1. 01

    Job

    The concrete thing to build. Tight enough to start in 72 hours.

  2. 02

    Why start

    One narrow job the market doesn’t serve yet — or the existing option is too expensive / too heavy.

  3. 03

    Evidence

    Primary links. Open them. No recap of a recap.

  4. 04

    72h path

    The first build slice. If that slice does not hold, stop.

Founding rate

First 50 founders

$19/month

$19/mo → 3 weekly “build this” picks. Not traffic. Pre-screened starts. First issue ships at 25 founders. You pay when checkout opens.

  • 3 weekly “build this” picks
  • Job, why-start, evidence, 72h path
  • $19/month locked for the first 50

Issue 1 unlocks at 25 founders 0 / 25

FAQ

Short answers.

When does the first issue ship?

When 25 founders are in.

What is a pick?

A start candidate: the job, why it is worth starting, evidence links, and a 72h path. Not a traffic plan.

Why only three?

Three is the product. A longer list is browsing.

Is this an idea list?

No. Each pick is pre-screened to start. Density is the point.

Who is this for?

English coding indies who already waste nights on HN, Reddit, and ChatGPT and have burned weeks on the wrong build.

Do you sell traffic or exposure?

No. Not traffic. Pre-screened starts.