What a Custom Web App Actually Costs in 2026 (€2,500 to €80,000+)
Most quotes for custom web applications are made up. Here is what real projects actually cost — from a €2,500 discovery sprint to an €80,000 enterprise platform — broken down by what drives the number.
Every custom web application quote is two numbers. The number on the proposal — and the number you actually pay six months later. They are not the same. They are rarely close.
I have built around fifty web applications over seventeen years — from €5,000 weekend MVPs to €80,000 enterprise platforms with 40+ data models and three years of follow-on maintenance. This piece is the cost breakdown I would give you if you called me before signing with anyone.
Why "it depends" is not an answer — and what it actually depends on
When you ask three agencies what a custom web application costs, you will get three answers that span an order of magnitude. €15,000 from one. €80,000 from another. €180,000 from a third. Same feature list. Same description.
That is not a bug. That is the market. The number depends on five things, and none of them are visible in a feature list:
- Who is doing the work (junior contractor vs. senior solo vs. agency team)
- What is in the contract (fixed price vs. hourly vs. milestone)
- How complete is the brief (clear scope vs. vague intent)
- What ships in v1 (everything vs. the minimum that proves the concept)
- What happens after launch (one-off vs. ongoing)
Every line of this article is about resolving those five variables to a number you can budget for. Skip them and you are estimating with vibes.
The honest answer is not a number. It is a range with reasons. Anyone who quotes you a single number from a one-paragraph brief is guessing.
Five honest cost ranges for 2026
These are the brackets I see and quote in actual 2026 projects. Numbers are EU freelance market with a senior solo developer. Agency multiplier is 2–3× for the same deliverable.
| Project type | Cost range | Timeline | Example |
|---|---|---|---|
| Discovery sprint | €2,500 | 1 week | Brief, wireframes, ADR, fixed-price proposal for build |
| MVP / Proof of Concept | €8,000 – €15,000 | 4–6 weeks | Landing + core feature + payments + auth |
| Standard business app | €15,000 – €40,000 | 8–14 weeks | Full app, 2–4 roles, 5–10 integrations |
| Complex business platform | €40,000 – €80,000 | 3–6 months | Multi-module, complex workflows, high availability |
| Enterprise / regulated | €80,000 – €150,000+ | 6–12 months | Regulated industry (FinTech, health), compliance audit, SLAs |
A note on the €2,500 discovery line — that is not a sales loss leader. It is the way I (and most serious senior contractors I respect) start client engagements in 2026. The sprint produces an artifact that has value even if the client never builds the project, which removes the awkward "free pitch deck" dance that wastes everyone's time.
What actually drives the cost
1 · Business logic complexity (the biggest factor)
A portfolio website with a contact form is not the same as a facility management portal with contract workflows, role-based permissions, and audit trails. The difference is not "more pages." The difference is the rules underneath.
Simple business logic (CRUD on a single resource, basic forms, list views): 60% of the cost goes to UI and design.
Complex business logic (multi-step workflows, approval chains, computed fields, conditional rules, time-based state machines): 70% of the cost goes to logic and 30% to UI.
For a facility management system I shipped recently, the project had four user roles, 22 data models, 45 permission rules, calendar scheduling with recurring patterns, and performance tracking with photo evidence. That level of business logic puts a project squarely in the €40,000–€60,000 range. The UI was deliberately minimal — every euro mattered more in the data layer.
2 · Number and quality of integrations
Every third-party integration has a real-world cost that depends entirely on the quality of the third party's API. Some integrations take a day. Some take a week.
Realistic time estimates from my last 20 projects:
- Stripe payments — 2–3 days for a clean checkout, 5–8 days if you need subscriptions, prorations, and webhooks
- Email service (Postmark, Resend, Mailgun) — 1 day for transactional, 3–4 days for marketing/templates
- AI/LLM integration (Claude, GPT) — 3–5 days for a clean wrapper, 2–3 weeks if it is a real agentic workflow
- CRM sync (HubSpot, Salesforce) — 3–7 days, depending entirely on whether you need real-time sync or batch
- Custom API integration — 2–5 days each, depending on the API's documentation and your appetite for retries
3 · User roles and permissions
A single-role application (every user sees the same thing) is straightforward. A multi-role application (admin, manager, employee, client — each with different data scopes and different actions allowed) multiplies complexity.
Every additional role adds 10–15% to development time, because every feature needs to be designed, tested, and audited from each role's perspective. A four-role system is roughly 1.5× the implementation cost of a single-role system with the same feature list. Plan accordingly.
4 · Design effort
Off-the-shelf component library (Tailwind + shadcn/ui or similar): minimal additional cost. The application looks competent within standard UX patterns.
Custom UI/UX design (Figma prototyping, custom components, distinct visual identity): adds €3,000–€10,000 to the project.
For most business applications I argue against the custom design unless the application is part of the brand promise (consumer-facing, design-led product, or marketing-critical surface). For internal tools and operational platforms, off-the-shelf design is almost always the right call. The money is better spent on UX clarity.
5 · What you do after launch
A web application is not a one-time purchase. It is a living thing that needs:
- Security patches when dependencies have CVEs (monthly to quarterly)
- Browser/OS compatibility updates as platforms evolve (yearly minimum)
- Small feature additions as you learn what users actually want (continuous)
- Infrastructure tuning as traffic grows (event-driven)
Budget 15–20% of the initial build cost per year for maintenance. A €30,000 application needs €4,500–€6,000/year in maintenance to stay healthy. Skip this and the app degrades — slowly at first, then suddenly — within 18 months.
Solo senior vs. agency: the real comparison
| Factor | Solo senior developer | Mid-size agency |
|---|---|---|
| Daily rate | €400–€600 | €800–€1,500 |
| Communication | Direct with engineer | Through PM layer |
| Consistency | One codebase style | Multiple devs, varying quality |
| Overhead | None (no sales, no office) | 30–40% of cost |
| Parallelism | Sequential work | Multiple workstreams possible |
| Typical MVP cost | €8,000–€15,000 | €20,000–€40,000 |
| Typical enterprise project | €40,000–€80,000 | €100,000–€200,000+ |
| Risk if person leaves | High (single dev) | Lower (team continuity) |
| Senior attention to detail | Always | Depends on PM and team mix |
The trade-off is real, not marketing. Solo senior gives you 40–60% lower cost, faster decisions, and consistent code — at the cost of parallelism and team continuity. Agency gives you parallel workstreams and continuity — at the cost of overhead and PM-mediated communication.
For most projects under €60,000 and shorter than four months, solo senior is the better economics. For projects above €100,000 or longer than six months with multiple concurrent workstreams, agency is the right structure.
Want a second opinion before signing a quote? That is what a discovery sprint is for — €2,500 and one week to make sure the build is scoped, priced, and contracted correctly.
How to budget — three rules I tell every client
Rule 1 · Start with the problem, not the feature list.
"We need a portal" is expensive to estimate. "We need to cut client onboarding from 5 days to 1 day" is specific enough to architect and price. The discovery sprint is exactly this translation: vague intent into a scoped brief.
Rule 2 · Ship the MVP, then learn.
Unless you are 100% certain about every feature, build the minimum viable product first. Launch with core functionality, get real user feedback, then iterate. I have watched too many projects burn €50,000+ building features that nobody used after launch. The cheapest feature is the one you decide not to build because the user data said it does not matter.
Rule 3 · Budget for year two.
A €30,000 build needs another €4,500–€6,000 in year-two maintenance. Add a small reserve (€3,000–€5,000) for the inevitable "one more feature" that emerges after launch. Total realistic year-one-plus budget for a €30,000 build: about €40,000.
Takeaways — what to ask before signing a quote
- Demand a written scope. Every feature, every role, every integration, on paper. Vague scope is how €15,000 projects become €40,000 projects.
- Ask who is doing the work. Senior solo, junior solo, agency team, or offshore. The same scope at different staffing levels has wildly different real cost (and wildly different real outcomes).
- Compare three quotes at the same level. Three agencies, OR three solo seniors, OR three offshore teams. Mixing apples and oranges gives you a useless range.
- Budget for year two. 15–20% of build cost, every year, for maintenance. If the quote does not include this conversation, the quote is incomplete.
- Start with a discovery sprint. €2,500 (one week) to get the scope locked, the architecture decided, and a fixed-price build proposal. Saves you €5,000–€30,000 in scope creep later.
Related: How to Hire a Full-Stack Developer in 2026 · Building Modern Web Applications · Services — Discovery Sprint
