The first version of a SaaS product has one job: get a small group of real users to do the main thing and pay for it, or tell you clearly why they won't. Most MVPs that drag on for months are trying to do more than that.
Write down the one action a user has to complete for the product to be worth anything. For a scheduling tool it's booking a slot. For an invoicing tool it's sending an invoice and getting paid. Everything in the first release should support that path, and almost everything else can wait.
Things that usually belong in version one: sign-up and login, the core workflow, billing if you plan to charge from day one, basic email notifications, and an admin view so you can see what users are doing and fix their data when something breaks.
Things that usually don't: a full settings area, team roles beyond owner and member, a mobile app, integrations nobody has asked for yet, and a custom design system. A clean standard UI is fine for the first users.
Pick boring tech. A mainstream framework, a managed Postgres database and a hosted auth provider will get you further than anything clever, and the next developer you hire will already know them.
Decide what you'll measure before launch. Sign-ups alone don't tell you much. Track how many people complete the core action in their first week and how many come back.
Plan the second month too. Budget time for fixing what the first users run into, because they will find things you didn't think of.
If you're deciding who should build it, I wrote up the questions worth asking an agency first: https://sitefusion.dev/blog/how-to-choose-an-ai-automation-agency
Our SaaS MVP work at SiteFusion: https://sitefusion.dev/services/saas-mvp-development
More on custom SaaS: https://sitefusion.dev/topics/custom-saas