I built a business that runs on two hours a week.

The constraint did the product management.

The rule was that it could never require full-time effort from anyone building it. Mine runs at two to three hours a week. That rules out a services business, any customer who needs hand-holding, and anything whose support load grows with installs.

What survives is narrow, self-serve and worth paying for on day one.

Shopify is the best place to sell anything until the thing is wine.

Checkout, customer record, apps, analytics, point of sale, all of it is there until you sell wine. Then alcohol has to clear the buyer's age, whether the destination state accepts the shipment at all, and that state's own fees. Every answer on the market solved that by sending the customer out to a compliance platform's checkout, or up to Shopify Plus at north of $2,000 a month.

Either way the winery loses the checkout, the customer record and its own data at the moment the sale happens.

We put compliance inside the merchant's own checkout. Age gate, per-state eligibility, and the state alcohol fee applied as a real line item, with the customer never leaving the store. Nobody had done that natively without forcing the merchant onto Plus.

I left the original prototype running below, live in a frame with every control still working. Take a look on the Shopify App Store.

Two buyers with nothing in common, so the docs became the funnel.

Research split the customer in half. One is the winery owner or the winemaker standing in a field, who is not technical and will never file a ticket. The other is the developer agency hired to build the store, reading our integration docs and never our end customer at all. They're a channel, and one agency relationship installs us across every winery they build.

So documentation became the go-to-market. 216 articles, published as a live docs site, written to answer the question before it gets asked, because one buyer won't ask and the other won't wait.

State fees and eligibility have to be right, and months of being right is what earned the club product access to their members' recurring billing.

I priced the first yes at $24.99 so the tenth one could pay.

Every product carries two plans. A low subscription with a higher cut of transactions, or a higher subscription with a lower cut. A winery starting out pays almost nothing to find out whether this works, and volume moves it up on its own arithmetic with no salesperson involved.

I took the cost knowingly. Revenue arrives slowly and stays small for a long time. Against a business that can never demand full-time effort, that is the right trade.

Clubs points the same thesis at the recurring revenue. Every existing option asked the winery to run its club somewhere else and hand the customer over again. Compliance gets a winery in the door first, and by then the data is flowing and the record already carries their eligibility, so Clubs costs almost nothing to acquire.

We built the club so a member picks their own six, sets the cadence, and pays through the winery's own checkout. See Clubs on the Shopify App Store.

Every channel had to keep working while nobody watched.

Paid acquisition needs a person watching it, so the rule ruled it out.

Where the installs come from

• The App Store listing. Written against what a winery types (wine club, age verification, ship wine), and carried by reviews.
• The docs site. Convincing before a conversation starts.
• The agencies. Each one we win installs us again on the next build.
• The compliance provider. Sending the wineries who already trust half the stack.

55+ wineries run their compliance through it. Compliance has been live on the Shopify App Store since May 2024 at 5.0 across eight reviews, and eight is a small sample. Clubs listed in 2026. Four of us, none full-time.

An agent pipeline does the work a product team would otherwise do. It is the same operating model I run at my day job, written up in How I build now.