The other articles in this guide give you fixed numbers for the common cases — a business website, an MVP, a redesign. This one zooms out and answers the bigger question underneath all of them: why the same brief can cost €500 or €150,000 — and how to land at the right end of that spread on purpose.
The uncomfortable truth first: the price of building for the web is not set by the technology. It's set by the approach you choose and the scope you commit to. A €500 project and a €150,000 project can start from the exact same sentence — "I need a place where customers can book and pay." The difference is what you decide that sentence has to actually do, and who you have build it.
The full price spectrum
Here is the whole ladder, honestly, with no tier hidden because it's inconvenient. Most guides written by people who sell at one tier quietly pretend the others don't exist. These are 2026 EU ballparks — read them as "which world am I in," not as quotes.
| Tier | Typical range | What it is | Best when | Main risk |
|---|---|---|---|---|
| DIY / no-code | €0–500 + monthly | You build it on Webflow, Framer, Tilda, Wix, Carrd | Validating an idea, a first landing page, a tiny budget, you enjoy tinkering | Your time; hits a ceiling fast; you rent the platform forever |
| Template customization | from €1,500 | A bought theme, restyled and filled with your content | You need "live and presentable" cheaply and the brand isn't the differentiator | Looks generic; weak SEO/performance out of the box; hard to extend |
| Custom fixed-price website | from €5,000 | Hand-built marketing site, custom design, SEO + speed baked in, CMS if needed | A serious business site that has to convert and rank | Almost none if scope is written down — this is the sweet spot for most businesses |
| Focused MVP (phased) | from €5,000 | One core flow built for real: auth, one payment path, live | You're testing whether people will pay for a product, not just read about it | Scope creep — the fix is to phase it, not to pre-buy everything |
| Product build | from €20,000 | Multiple roles, dashboards, real integrations, real-time, heavier logic | The core flow already has paying users and demand is proven | Building for imagined demand instead of measured demand |
| Enterprise / regulated | from €150,000 | A team, compliance (fintech, health), high scale, SLAs, audits | Real regulatory or scale requirements — not vanity | Paying team prices for a problem that a focused build would solve |
Two things this table is meant to kill:
- The idea that "a website" is one price. It's six different products wearing one word.
- The idea — which the honesty of this guide can accidentally imply — that anything above €20,000 is just padding. It isn't. Real product work genuinely costs five and six figures, for real reasons, and pretending otherwise would be its own kind of dishonesty. The next two sections separate the legitimate reasons from the padding.
What this looks like in practice
Concrete scenarios across the spectrum — same anchors as the table above, with a rough timeline so you can find the one that looks like your project:
| Project | What's in scope | Timeline | Price |
|---|---|---|---|
| Landing / micro-site | One page, your content, a real payment button — built to validate demand fast | Days–1 week | from €1,500 |
| Business website | 8–10 pages, custom design, SEO and speed baked in, CMS | 2–3 weeks | from €5,000 |
| Focused MVP | One core flow done for real: auth, one payment path, live and measurable | 2–4 weeks, phased | from €5,000 |
| Real product | Marketplace or portal: multiple roles, real integrations, dashboards | 2–4 months, phased | from €20,000 |
| Enterprise / regulated | A team, compliance, scale, SLAs | Months | from €150,000 |
These are senior, one-expert prices: you work directly with the person who designs and builds the thing. They sit deliberately above the offshore and marketplace floor — I don't compete on being the cheapest, I compete on getting it right the first time, so it doesn't get rebuilt in a year.
What actually drives the number
Within any tier, and when moving between tiers, these are the levers that move the price — roughly in order of impact:
- Scope definition. By far the biggest. "Users book appointments" is one sentence and ten different products depending on the answers to: which auth, which payments, how many roles, what admin, which integrations. Every "yes" is days of work. This is why a vague brief is expensive: it silently assumes the maximum.
- User roles and permissions. One kind of user is simple. Customers + providers + admins, each with a different view and different rules, is not "three times" the work — it's the work, multiplied by every screen. Marketplaces live and die here.
- Integrations. Each connection (payments, CRM, calendar, SMS, accounting, an old internal system) is a small project. Five of them together are a meaningful part of a build, and the undocumented or legacy ones can eat days each.
- Data, security and compliance. Storing an email is trivial. Handling payments, health data, or anything under GDPR/PCI/regulatory rules adds real, non-negotiable engineering — and it's exactly where cheap builds cut corners you can't see until they cost you.
- Real-time and scale. Chat, live updates, notifications, thousands of concurrent users — each pushes you up a tier because the boring infrastructure underneath gets genuinely harder.
- Custom logic. A matching algorithm, a scheduling engine, a pricing calculator with edge cases. This is product engineering, not page-building, and it's where product tiers earn their number.
- Unique page designs. For sites specifically: three templates reused across ten pages is cheap; ten bespoke layouts is not. Design is where senior time goes.
- Content, languages, and who supplies them. Ready content = assembly. Missing content = copywriting, sourcing, strategy. Each extra language multiplies content work and adds technical SEO (hreflang, localized metadata).
- Who builds it. One senior person carries no account-manager, project-manager, or sales overhead. A team carries all three — sometimes necessary, often not. This is the single biggest reason two quotes for the "same" thing differ by 3×.
Why the top of the range is real — not just overhead
It would be easy to read the rest of this guide and conclude that everything expensive is agency padding. Some of it is. But a lot of it isn't, and you should be able to tell them apart before you negotiate.
Legitimately expensive (the money buys something you can't skip):
- A two-sided marketplace where supply and demand each need their own onboarding, dashboards and trust mechanics.
- Anything touching money at scale, or regulated data (fintech, health, insurance) — the compliance and security work is real and slow.
- Real-time systems, or products that must stay up under real load, with on-call and SLAs.
- Genuinely novel logic — the thing that doesn't exist yet and can't be bought off the shelf.
Padding (the money buys structure, not outcome):
- Layers of managers between you and the person writing the code.
- A team-sized proposal for a 6–10 page marketing site.
- "Enterprise" framing on a problem that a focused, phased build would solve for a fraction.
The test: ask what specifically in the price requires a team rather than a senior individual. Legitimate answers are concrete (roles, compliance, scale, 24/7). Vague answers ("that's just what it costs at this level") are the padding talking.
How to spend less — without wrecking the result
Cutting cost well is not about haggling the rate. It's about buying the same outcome with less scope, in the right order. The levers that actually work:
- Build for your riskiest assumption first. For most products the risk isn't "can this be built," it's "will anyone pay." A landing page with a real payment button tests that in days for the price of a template. Only build the platform once the core flow has paying users. This one decision moves cost by multiples — more than any discount ever will.
- Phase everything large. Never pre-buy month four on a month-zero guess. Ship the riskiest, most valuable slice; scope the next phase only once the last one is live and teaching you something.
- Reuse where it doesn't sell, invest where it does. Bespoke design on the pages that convert; reused templates on the ones that don't. The same logic applies to features — build custom only where it's your differentiator.
- Prefer standard patterns and off-the-shelf building blocks. Stripe for payments, a proven auth provider, a headless CMS — not a hand-rolled version of each. "Standard" is why a fixed price is even possible; bespoke everything is how budgets detonate.
- Start no-code to validate, rebuild what earns. An AI prototype or a no-code build is a cheap, smart way to test interest. Just treat it as validation spend, not build spend — and rebuild the parts that have to carry real users and real money before they break in public.
- Bring your own content, ready. "We'll send the content next week" is the number-one reason projects run over. Content that exists and is good turns writing into assembly.
- Match the builder to the size of the job. A 6–10 page site rarely needs a full team; a regulated fintech platform does. Paying team prices for solo-sized work — or solo prices for team-sized work — both end badly.
- Insist on fixed price and written scope. This is a cost-control tool, not just a comfort. It moves the risk of a bad estimate onto the builder and makes scope creep a visible, agreed decision instead of a silent invoice.
Where cutting costs backfires — the false economies
Some savings are real. These four are borrowed money at a bad interest rate:
- SEO structure. A site invisible to search engines gets rebuilt or re-fixed, usually for more than it saved. Non-negotiable from day one.
- Performance. A slow site loses visitors and ranks worse. The discount gets repaid in lost customers, quietly, forever.
- Ownership. If you don't own the code, domain and accounts, the low price came with a leash. Leaving costs you the whole product.
- Security, on anything with real users or money. The corner you can't see until it's a breach. Cheap here is the most expensive line on the list.
Spend less by narrowing scope and sequencing it well — never by skipping the four things above.
The one-sentence version
Decide which tier your idea actually needs (not the one you're afraid of, and not the one you're hoping for), build the smallest slice that tests your riskiest assumption, phase the rest, and refuse to save money on SEO, speed, ownership or security. Do that and you'll spend at the bottom of whatever range is honestly yours.
If you want that mapped onto your idea — which tier, what to build first, and a fixed number for it — book a free 20-minute call. You'll leave with a straight answer, even if the honest one is "start cheaper."