How much does a SaaS MVP cost in 2026?
"SaaS MVP cost" is where founder conversations get uncomfortable fast. One agency quotes €12,000, another €85,000 — for the same three-paragraph brief. The gap is real and usually explained by what each party means by "MVP." Below are the price bands we use for MVP and startup development, with an honest map of prototype vs no-code vs custom code — and what to cut before you pay for scale you do not need yet.
Price bands by build path
| Path | Cost (EUR) | Timeline | Best for |
|---|---|---|---|
| Figma prototype | €2,000 – €5,000 | 2–4 weeks | Idea validation, investor demos, user tests |
| No-code MVP (Bubble, etc.) | €8,000 – €18,000 | 4–8 weeks | Simple workflows, <500 users, fast experiments |
| Custom code MVP | €18,000 – €45,000 | 8–14 weeks | Auth, Stripe billing, one core workflow, clean IP |
| Production v1 | €45,000 – €120,000 | 4–8 months | Multi-role admin, analytics, SLA, security hardening |
What actually moves the number
- Scope clarity. "User management + dashboard + API" can mean 80 or 800 hours. Written discovery (€500–€2,000, often credited) removes guesswork.
- Billing complexity. Single Stripe Checkout: modest. Usage-based metering, seat pricing, trials, and proration: +30–50%.
- Real-time features. Live collaboration, websockets, notifications at scale — rarely needed in a true MVP.
- Mobile. Responsive web first; native iOS/Android is a separate product line, not an MVP default.
The mistake we see most often
Founders skip validation and fund a €40,000 "MVP" with admin panels, role matrices, and microservices — then discover the core workflow was wrong. A €5,000 prototype plus 15 customer calls would have changed the spec entirely.
Read also: MVP vs prototype — which fits your stage.
No-code vs custom code — when each makes sense
Bubble, Glide, Webflow + Memberstack — they can raise a functional MVP in 4–8 weeks for linear workflows: form → approval → notification → export. The limitation appears when you need: async background jobs, integrations with APIs requiring complex transformations, or fine-grained control over security and permissions. No-code also taxes you at scale: Pro plans quickly reach €500–€1,000/month when you exceed 1,000 active users, and data migration out of the platform is painful.
Custom code (Next.js + Node.js/Python + PostgreSQL) has a higher initial cost (€18,000 vs €8,000), but the IP is entirely yours, you can self-host on a €15/month Hetzner server, and any senior developer can continue the project. For founders seeking investment, custom code with full ownership is a positive signal in due diligence.
What a properly built SaaS MVP includes
- Authentication. Email + password (with reset flow), optional Google OAuth. Session management with JWT or httpOnly cookie. Do not use localStorage for tokens.
- One core workflow. The single thing users do 80% of the time. Everything else waits for feedback.
- Billing. Stripe Checkout (simpler) or Stripe Billing with webhooks for subscriptions. Usage-based pricing needs custom metering — adds complexity.
- Minimal admin. View users, deactivate accounts, see synced Stripe transactions. Do not build a CRM into your MVP.
- Basic analytics. PostHog or Plausible for funnels and retention. Not Google Analytics — it does not tell you which features users engage with.
Realistic timeline and what happens after launch
A custom-code MVP takes 8–14 weeks: weeks 1–2 discovery and architecture, weeks 3–8 implementing the core workflow with auth and billing, weeks 9–12 QA, bug fixing, and onboarding the first 10–20 users. The first 30 days post-launch are critical: you measure activation rate (how many users complete the key action in their first session), 7-day retention, and collect qualitative feedback through 15-minute interviews.
Budget a post-launch iteration (€3,000–€8,000, 2–4 weeks) for data-driven adjustments. The most common scenario: the core workflow works, but onboarding is too long — users abandon before seeing value. Reducing from 7 steps to 3 increases activation by 20–40%.
Case study: B2B SaaS MVP
A client wanted a contract management platform for law firms (15 lawyers, 200 active contracts). The initial discussion sketched a full CRM with calendar, invoicing, time tracking, client portal, and custom reports. The competing agency estimate: €65,000, 6 months. After discovery (2 weeks, €3,000), we identified that 80% of the value was: upload document, assign lawyer, deadline reminder, and PDF export. We delivered the MVP in 10 weeks for €22,000 — remaining features would be evaluated after 60 days of real usage.
After 90 days, the client requested only two new features: tags on contracts and full-text search. No lawyer used the calendar or time tracking from the initial spec — confirming that the scope cut in discovery was correct. Savings: €43,000 and 4 months that would have been invested in unused features.
This is the rule, not the exception. In 80% of the SaaS projects we ship, the first iteration discovers that half the planned features go unused. The MVP is not a cheaper version of the final product — it is the learning tool that tells you what to build next.