← The Field
Workflows 5 min read

How I shipped my first $49 product by treating the launch as a checklist, not a department

I had a finished product sitting in a folder for two months because the launch infrastructure felt like a team-sized job. It took three evenings and about four hours of real build time, plus a one-dollar test purchase that caught what I'd never have seen.

TL;DR

I had a finished product sitting in a folder for weeks because the launch infrastructure (storefront, checkout, fulfillment, payments) felt like a team-sized job. It took three evenings and about four hours of actual build time. This post documents the pipeline and the verification step that saved me from a real embarrassment. 6 minute read.

The Problem

The product was done. A 16-page kit with three template files, zipped and ready. What I didn’t have was everything around it: a page to sell it from, a checkout that takes cards, a way to deliver the file, and proof that money could actually move from a stranger’s card to my bank account.

That list is why solo operators sit on finished work. None of it is hard, exactly. It’s that each piece belongs to a different job description: web developer, payments admin, fulfillment ops. When you’re the whole department, “I’ll launch when the infrastructure is ready” quietly becomes two months of not launching. I know because my email list had two subscribers at launch, and one of them is me.

The Workflow

Six steps, in the order they actually happened.

  1. Product file to sale-ready PDF (Claude). Took the working docx and turned it into a 16-page sale-ready PDF, plus three companion template files, bundled as a zip. This also caught a leftover reference to an offer I’d retired, which would have shipped a dead link to every buyer.

  2. Storefront (Claude Code). Built a kit catalog page and a product detail page on the site, with redirects from the old URLs. I described what I wanted; the AI wrote the pages in the site’s existing design system.

  3. Checkout (ThriveCart). One product, $49, one-time. Pointed it at a custom subdomain so the checkout URL looks like mine and not like a cart vendor’s.

  4. Fulfillment (Claude Code). A branded download page on my own site, with the zip delivered from there. ThriveCart’s success page redirects there after purchase.

  5. Verification (me, $1). Set the price to $1, bought my own product live with a real card, and followed every link a customer would touch: checkout, redirect, download, receipt email. Then reset the price. More on why this step earns its keep below.

  6. Feedback loop (Claude Code). A short feedback form linked from the download page, asking one required question: which workflow should go deeper. Responses go to a database and my inbox. This form is the research mechanism for what I build next.

Steps 2, 3, and 4, the parts that felt like a team-sized job, took about four hours total.

What Broke

Three things, in escalating order of “glad I caught that.”

The site kept saying “coming soon” after the product went live. The storefront page needed a redeploy to flip the status. For a stretch, the product was purchasable but the shelf looked empty. Live product, invisible product.

The dollar went missing. My $1 verification payment didn’t show up where I was looking. I spent a stretch of launch night hunting through two payment dashboards before finding it: right account, wrong screen, plus a date filter quietly hiding the transaction. Nothing was broken. I just didn’t know my own payment plumbing yet, which is its own finding.

The big one: my first test purchase landed on a page of placeholder Latin. The cart’s default confirmation page, lorem ipsum and all. That would have been a real buyer’s first impression of the brand. The fix was a single buried toggle (“automatically redirect to success URL”). The only reason a customer never saw it is that the first customer was me, on purpose.

The Result

A live, purchasable $49 product, verified end-to-end with a real transaction on a real card. Storefront, checkout, fulfillment, and feedback loop, built in three evenings around a day job, with roughly four hours of hands-on infrastructure work.

There’s no honest before-and-after time comparison here, because before AI I would not have built this at all. The real ledger reads: two months of a finished product sitting in a folder, then three evenings once I treated the infrastructure as a checklist instead of a department.

The Prompt

The most valuable artifact from launch week isn’t a prompt. It’s the verification checklist. Steal it:

LAUNCH VERIFICATION (run in live mode, not test mode)

1. Drop the price to $1.
2. Buy your own product with a real card, in a private browser window.
3. Follow every link a buyer touches:
   - checkout page (does it look like your brand or the cart vendor's?)
   - post-purchase redirect (did you land on YOUR page, or placeholder text?)
   - the download itself (does the file open, is it the current version?)
   - the receipt email (click every link in it)
4. Find your dollar. Trace the payment to the exact dashboard screen
   where it lives, and confirm where and when it pays out.
5. Check your own site's shelf. Is the product visible and marked available?
6. Reset the price. Keep the $1 receipt.

Every broken thing I found, this checklist found. Nothing else would have.

Take It Further

The packaged version of the workflows I actually sell is in the kit catalog at solobuilt.ai/kits. And every workflow I document lands in The Dispatch first, Tuesdays at 6am ET: solobuilt.ai.

The dispatch

One workflow, every Tuesday morning.

Be among the first subscribers. Real workflows from one person doing the work of a whole department. Free, forever.

No tracking pixels. No drip campaigns. Unsubscribe anytime.