A men's coaching membership, off six subscriptions in eleven weeks.

The situation

Six tools, one business, nobody in charge of the seams.

The program had grown the way most do: a tool for each new need, added when the need arrived. By the time we met, members had multiple logins, the team was copying information between systems by hand, and the founder had a list of things he wanted the platform to do that the platform had been promising for eighteen months.

The monthly software bill was $840, before the annual increases. More expensive than the bill was what it prevented: offers that couldn't be launched, member experiences that couldn't be improved, and a business designed around what six vendors happened to allow.

The decision

Own the layer that makes the business different.

The Stack Audit put every subscription, workflow and limitation in one document and did the arithmetic. The exit plan kept the commodities — payments stayed on Stripe, calls stayed on Zoom — and replaced the custom layer: the member experience, the coaching operations, the content, the community and the back office.

The build was quoted at a fixed price. Break-even against the old bill landed at month 14.

The build

One system, one login, everything the program does.

The system that replaced the six subscriptions runs the whole membership: a member dashboard, coaching and session booking, live meetings with archives, private messaging, education tracks, podcasts and audiobooks, a members map, check-ins, notifications, and the admin back office the team runs it from.

It was built on plain PHP and MySQL on ordinary shared hosting in the client's own name, in the client's own repository, with a runbook and a recorded walkthrough delivered at launch.

The migration

Eleven weeks, nothing went dark.

Members, purchases, subscription states and content were exported, cleaned and imported. Old and new ran in parallel until the last member had logged into the new system. The six subscriptions were cancelled in the order the exit plan set out — the last of them only after the parallel period ended.

Afterwards

The roadmap became a to-do list.

Every feature request that had been sitting open since 2024 was shipped within six weeks of launch. The one the founder had waited eighteen months for took four days.

The program stays on a Care & Growth Plan — maintenance, a steady stream of improvements, and a monthly working session on the numbers. It could cancel any month and keep everything. It hasn't.