Afsar Ali · 4 min read
Founders usually arrive with a feature list. The list is the problem.
An MVP exists to answer one question: will someone use this, and will they pay? Every feature that does not help answer that question is deferred spend on an idea nobody has validated yet.
Start from the single core action
Name the one thing a user must be able to do. Book the appointment. Place the order. Generate the report. That action, done well end to end, is the product. Everything else is packaging.
What almost always gets cut
Admin panels before there is data to administer. Role systems before there is a second user type. Dark mode. Notification preferences. Multi-language before the first market works. Analytics dashboards measuring traffic that does not exist yet.
What almost never gets cut
Authentication that works. A payment path if you intend to charge. A data model you will not need to throw away in month three. Basic error handling, because a broken first impression is expensive.
The test
If you cannot describe the MVP in two sentences, it is not an MVP yet. Cut until you can, launch, then let real usage decide what comes next.