How Much Does It Cost to Build a SaaS MVP?

The €6,000–22,000 band is wide because two MVPs with identical pitches can carry very different engineering loads. The single biggest swing is billing logic. A flat monthly plan with one Stripe price is a day of work; trials, annual vs. monthly toggles, proration, metered usage, and reliable webhook handling (so a failed charge actually downgrades the account) can be a week-plus on their own.

The second swing is multi-tenancy. Row-level security on a shared schema lands near the low end. Separate databases per tenant, or custom data-isolation rules, push toward the high end fast.

Third is how many account types exist. One user role is cheap. Add an admin who manages other users, an org owner who invites teammates, and read-only viewers, and you have multiplied the permission matrix that every screen has to respect.

To stay near €6k, cut to one tenant model (shared schema + row-level security), one paid plan with a trial, email + one OAuth provider, and a single core feature loop — the one thing the MVP exists to prove. Skip the admin portal; a database client or a simple internal query gets you through the first months.

What pushes you toward €22k, in rough order of cost:

• Admin tooling — dashboards, usage analytics, and internal control panels often cost as much as the user-facing app, sometimes more.
• Auth depth — SSO, magic links, and granular role-based permissions each add real time.
• Metered or usage-based billing instead of flat plans.
• Third-party integrations beyond a basic webhook.

A useful rule: anything that only your team sees (admin panels, ops tooling) is the easiest scope to defer without hurting the validation goal. Build it once you have paying users telling you what they need.

The build quote is not the whole number. Recurring costs that rarely make it into early budgets:

• Stripe fees — roughly 1.5–3.4% + a fixed fee per transaction depending on card region. This scales with revenue, not a one-time line.
• Hosting — a Vercel-style deploy plus a managed database and auth provider commonly runs in the tens of euros per month early on, climbing with traffic and storage.
• Transactional email — password resets, receipts, and invites need a sending service; free tiers run out as you grow.
• Domain, monitoring, error tracking — small individually, real in aggregate.
• Maintenance — dependency updates, security patches, and Stripe/API changes don't stop after launch. Budget for ongoing work, not a one-and-done.

Mobile apps are also a separate scope, not a cheap add-on — and shipping to the App Store carries Apple's annual developer fee plus review overhead.

For a SaaS MVP, fixed-scope fits best because the deliverable is definable up front: this feature loop, this billing setup, this auth. Fixed scope means you know the number before work starts and the risk of estimate drift sits with me, not you. Hourly only makes sense once the product is live and you're iterating on open-ended changes where the next task genuinely can't be specified ahead of time.

The honest part of fixed-scope: it requires a locked spec. If the feature set is still moving, lock the v1, ship it, then price the next slice separately.

Realistic timeline is 6 to 14 weeks solo, tracking the same drivers as cost — lean MVP near six weeks, admin-heavy or multi-role builds toward fourteen. For reference, BookBed (multi-tenant booking SaaS with bidirectional iCal sync, on Flutter + Firebase + Stripe) and Callidus (clinic SaaS with row-level security, on React + Firebase) both went from zero to live production in under 12 weeks. Solo work runs leaner than an agency on this scope, because there's no layer of account management or internal handoffs between you and the code.