Website, platform or MVP: which of the three do you actually need
Most first briefs describe an MVP and mean a website. A short, honest way to work out which of the three shapes your project actually is before anyone quotes for it.
The three words get used interchangeably, and they should not be. A website, a platform and an MVP are different products with different costs, different timelines and different failure modes. Getting the classification right before a proposal is written saves months and, usually, a lot of money.
A website is a considered public surface#
It exists to tell a specific audience who you are, what you do and what happens next. There may be a form, a booking flow or a private area, but the site itself is mostly editorial. If the honest description of the project is 'the current site is embarrassing and we need it to reflect who we are now', this is a website.
A platform is a private product with real users, real data and a running back office. Members log in. Roles differ. Content or transactions flow through it. There is an admin surface behind the public surface. A directory, a portal, a members' area or a booking system with real operations behind it is a platform, not a website, even if it looks similar from the outside.
An MVP is the first serious version of a product that does not exist yet, built to test whether a specific audience will actually use it. It is smaller than the founder's ambition on purpose. It exists to make one bet, cheaply, in front of real users, before more money is spent. If the answer to 'what happens if nobody uses this' is 'we learn something and adjust', it is genuinely an MVP.
The most common mistake is scoping a platform as if it were a website#
The visible pages get built cleanly and cheaply, and then the hidden work, the roles, the admin, the data model, the edge cases, arrives as a surprise halfway through. The second most common mistake is scoping an MVP as if it were a finished platform, spending twelve months and a large budget on features that could have been tested in six weeks.
There is a simple test#
Draw the product on one page. If everything on the page is content someone reads, it is a website. If part of the page is a private surface with logged-in users doing something, it is a platform. If the whole point is to find out whether the private surface should exist at all, it is an MVP.
Once the shape is honest, the timeline, budget and team all follow naturally. Skipping the classification is where projects quietly lose their first six weeks.
Want the full website brief guide?
We've turned this process into a free, founder-friendly guide with templates and examples.
Request private consultation →Related articles
SEO that holds up in 2026: the small number of things that matter
Most SEO advice for 2026 is either panic about AI answers or the same technical checklist recycled since 2015. A calmer view of what a serious site actually needs to do.
Read →AI tools worth keeping in 2026: a short, honest list
Most AI tool lists are affiliate round-ups. This is a short account of what a considered team actually uses week to week, and why the list is shorter than the industry wants it to be.
Read →