What a Website Really Costs — The Anatomy of a Budget

I have quoted hundreds of projects. The first question, almost every time, is how much does a website cost. It is a fair question and I wish I had a number for it. The trouble is that it has exactly the same shape as how much does a house cost: the answer depends on the plot, the floor area, the finishes, who builds it, and how fast you want to move in.

The difference is that with houses, everyone intuitively accepts that a studio on the outskirts and a villa with a pool are not comparable. With websites, many people assume there is a standard product at a standard price, and that price differences are just greed or naivety. They are not. They are differences in the actual content of the work.

A price without a defined scope is not a quote. It is a guessing game that both parties play, and whoever guesses badly pays for it later.

When I break an estimate apart, nearly the whole budget settles into seven piles:

• Scope. How many page types exist, not how many pages. Ten pages sharing one template cost about as much as one. Three pages with three different structures cost three times as much.
• Content. Copy, photography, translations. If it arrives from the client, already written, a serious chunk of effort disappears. If I write it, that is a separate craft and it is billed as one.
• Integrations. Payments, CRM, ERP, bookings, invoicing, newsletter. Every external system adds documentation to read, failure cases to handle, and tests to write. Integrations are where estimates break most often.
• Custom design versus template. A bought template costs little up front and a lot every time you deviate from it. A design built for your brand costs more up front and almost nothing on deviation, because there is nothing to deviate from.
• States and edge cases. A screen is not a picture. It is the empty state, the loading state, the error state, the label that is three times longer than in the mockup, the old phone, the bad connection. Half of development time lives here, and this is exactly where a site that photographs well diverges from a site that works.
• Revisions. Two structured rounds are healthy. Unlimited revisions are not generosity. They are a way of moving risk into either the initial price or the final quality.
• Timeline pressure. Urgency is a cost function, not a preference. Two weeks instead of six means people working in parallel, overlap, and rework.

This is where clients lose the most money, and never on the day they sign.

• Copy and photography. A good site with badly chosen stock images looks like every other site. A photo session and a copywriter are real budget lines, not optional extras.
• Hosting, domain, certificates. Small monthly, permanent forever.
• Maintenance. Security updates, dependencies, vendor API changes. A website is software, and software rots if nobody touches it.
• Third-party subscriptions. Forms, maps, advanced analytics, search, transactional email, payments. Individually trivial. Added up, a monthly invoice nobody planned for.
• Your own time. Feedback, approvals, content, decisions. Stalled projects are rarely stalled on the developer side. I wrote separately about why I take one project at a time — attention has a price.

A cheap website is not cheap. It is deferred. Three invoices arrive later.

The first is the rebuild. A site rushed onto weak foundations asks to be rebuilt within twelve to eighteen months. You paid for two websites and got one.

The second is lost conversions. If the site loads slowly, has a clumsy form, or fails to inspire trust on mobile, you lose customers every single day, silently. It appears on no invoice, and it is the largest cost of the three.

The third is SEO debt. Wrong URL structure, duplicate content, missing redirects, poor speed. It accumulates for months and is repaid with interest at migration time. I covered the wider context in my read on the Romanian web market.

Hourly is honest when scope is unclear or evolving — maintenance, research, iteration. The risk sits with the client, so it only works with a provider you trust and an agreed ceiling.

Fixed fits when scope is defined. Risk moves to the provider, and the provider prices that risk in. An honest fixed price is always slightly higher than a perfect hourly estimate, and that is fine: you are buying predictability.

Value based makes sense when the site is a measurable sales channel. It is not for everyone and it is not an excuse to triple the number, but for a shop or a portal with direct revenue it is the most honest conversation available.

You do not need a forty-page specification. You need one page answering six things: what the business does, what a visitor must be able to do on the site, what page types you anticipate, which external systems connect, who supplies copy and images, and what budget and deadline you have in mind. Naming a budget does not weaken your position. It lets the provider design the right solution, exactly as you tell an architect what you intend to invest before they start drawing.

From the provider: a price given in five minutes with no questions asked, a one-page quote with no written scope, a promise of unlimited revisions, refusal to show real work, silence when you ask who owns the code and the accounts. From the client: refusal to name a budget, comparing quotes that describe different things, insisting on starting with no content ready, and most commonly of all, scope that grows after signature under the phrase let us just add one small thing.

Scope written in sentences, not vague bullets. What is included and, just as importantly, what is not. The assumptions the price rests on. Milestones and payments. The number of revision rounds. Who delivers content. What recurring costs look like after launch. Ownership of code, domain, and accounts. And a definition of done, meaning the criteria by which the work is accepted.

If you want to see scope translated into finished work, real projects explain it faster than descriptions do. And if you already have that one-page brief written, send it over — a price quoted against a clear brief is the only price that survives to the end of the project.