SaaS MVP: what to build first, and what to leave out

The four things that cannot wait

1. One workflow, done properly

Pick the single job your customer hires the product to do, and make that path excellent. Not three jobs done adequately. If someone cannot complete that one workflow and get value from it in the first session, no amount of surrounding feature work rescues the release.

2. Auth with tenancy designed in

Accounts, roles and tenant isolation belong in the first release, not because early customers demand them but because adding them later means touching every query in the system. This is the one piece of architecture worth over-investing in on day one.

3. A way to take money

Free pilots teach you almost nothing about willingness to pay. Subscriptions, trials, failed-payment handling and a basic invoice, wired to Stripe or Razorpay, belong in v1 — the moment someone enters a card is the moment your feedback becomes trustworthy.

4. Enough admin tooling to support a customer

You will need to look at an account, fix a bad record, extend a trial and see what a user actually did. Without it you will be running database queries in production at 11pm, which is a worse use of your first month than any feature you deferred.

What can wait, almost always

• Integrations nobody has asked for by name. Build the one your first five customers all mention; ignore the roadmap of thirty logos.
• An analytics dashboard for your users. Early customers want the job done, not charts about the job.
• A settings page. Every toggle is a branch to support forever. Pick opinionated defaults.
• A mobile app. A responsive web app answers this until usage proves otherwise.
• Onboarding tours and empty-state illustrations. Talk to your first twenty users personally instead.
• Enterprise features — SSO, audit logs, custom contracts — until an enterprise deal is actually on the table. Then they are worth building properly.

How to phase it

1. Weeks 1–2: architecture, auth, tenancy, the skeleton of the core workflow.
2. Weeks 3–5: the core workflow end to end, with real data and the admin tooling to support it.
3. Weeks 6–7: billing, trials, the signup path and the emails that go with them.
4. Week 8 onward: put it in front of paying users and let what they do decide the next phase.

We quote platform builds phase by phase for exactly this reason: a fixed price on a full platform is either padded or wrong, while a fixed price on the first usable release is something both sides can hold to.

The test for whether your MVP is small enough

Write down what a customer must be able to do for the product to be worth paying for. If that sentence needs the word 'and' more than once, the release is too big. Cut until it does not, ship, and add the rest with users watching.