The four-week MVP playbook
How we take a napkin sketch to a launched product in four weeks, the process, the cuts, and the traps.
Four weeks is not a marketing number. It's the shape of a well-run 0→1 build when a senior team owns the whole thing and the founder makes decisions on time. Here's the actual playbook.
Week 0, Discovery (before the clock starts)
Nothing gets built until we can write the brief back to you better than you gave it to us. One working session, then a written scope: the one thing this MVP must prove, the single user it's for, and the ruthless list of what we're not building yet.
The most valuable output of week 0 is the cut list. Every feature you name gets sorted into "week one," "later," or "never." Founders who protect the cut list ship in four weeks. Founders who relitigate it every Monday ship in twelve.
Week 1, Spine
Design system and the core flow, end to end, even if half the screens are placeholder. Auth, data model, one real path a user can walk. You should be able to click through the skeleton of your product by Friday. Ugly is fine. Broken is not.
Week 2, Muscle
The feature that makes it your product, not a template. This is where the AI, the real-time layer, the payment flow, or whatever the actual point is gets built and wired. Weekly demo, and we adjust the cut list based on what we've learned, always cutting, never adding without cutting something else.
Week 3, Skin
Polish, states, edge cases, the empty screens, the error screens, the "you have no data yet" screens that make software feel finished. Performance pass. Mobile pass. The details that separate "a dev built this" from "this is a product."
Week 4, Ship
QA, deploy, docs, and the launch surface, a landing page and the messaging that goes with it, because we're not just handing you a repo and disappearing. You go live with users able to transact.
The traps
- Scope creep disguised as feedback. "Can it also…" is the four-week killer. It goes on the later list.
- Absent decision-makers. We can build fast. We can't decide for you. A founder who answers within a day is worth more than one more engineer.
- Perfectionism on the wrong things. Polish the flow the user actually walks. Nobody cares about the settings page yet.
Four weeks gets you a real, launched thing you can put in front of users and learn from. That's the whole game at this stage, not a perfect product, a learning product, live.
Want yours in four weeks? Book a call.
