Fig. 14 — First version
MVP development
Four to twelve weeks from a first conversation to something real people are using. Not a prototype, not a demo — a product that is deployed, watched, and safe to change.
What four to twelve weeks buys
The spread is wide because the work genuinely is: four weeks when the problem is narrow and the decisions are already made, twelve when discovery has real work to do first. What ships at the end is the same either way — a running product with users on it, not a handover document.
- A first version narrow enough to finish and complete enough to judge
- Deployed to a real environment from the first week, not at the end
- The data model designed to survive the second version, because it rarely survives being an afterthought
- Something you can put in front of a customer and learn from
Shipped, not demoed
Most of what separates an MVP from a prototype happens after the feature work is done. We treat that part as included: the pipeline refuses to ship a build that fails typecheck, lint, tests or the end-to-end suite; staging mirrors production rather than approximating it; database migrations run through a migration tool, never by hand against live data; and health checks and rollback exist before the first incident rather than after it.
When an MVP is the wrong thing to buy
An MVP answers a question you do not yet know the answer to: will anyone use this. If you already know exactly what has to exist — because the process runs on paper today, or because a system you depend on has to be replaced — then "minimum viable" is the wrong shape and a narrow first version only delays the thing you actually need. That is a different conversation about scope and budget, and worth having on the call rather than discovering three months in.
Scope is the lever, and it is yours
The date moves or the scope does, and we would rather cut scope with you in week two than discover the problem in week ten. Every commercial model works on our side — fixed for a defined scope, time and materials, or a monthly rate — so the shape of the contract can follow the shape of the risk instead of fighting it.