You ask five different places to quote for a website. You send all of them the same text. The proposals come back ranging from a few thousand to five times that. The natural reaction is suspicion: someone is overcharging, or someone is undercharging and will turn up later with extras.
Sometimes that's exactly what's happening. Most of the time it isn't. The spread exists because «a website» isn't a defined product, and your brief — however carefully you wrote it — leaves dozens of decisions open. Each proposal filled those gaps its own way, based on what each studio assumes by default. You're comparing different things that happen to share a name.
This article is here so you can understand what sits inside each number, and read the proposals instead of sorting them by price.
The brief you sent isn't enough
Picture a typical request: «we need a new website for our company, around eight pages, services, team, blog and a contact form». It sounds specific. It isn't.
It doesn't say who writes the text. It doesn't say whether photographs exist or have to be made. It doesn't say whether there's an established visual identity, or whether the website will be the first place the brand appears properly designed. It doesn't say whether the site is in one language or two. It doesn't say who updates the blog, or how often. It doesn't say whether existing content migrates or whether you're starting from scratch.
Every one of those questions moves the figure by thousands. And because most quote requests don't answer them, every studio assumes. Some assume the minimum scenario so they can present an attractive number. Others assume the full scenario because they know that's where the project will end up anyway.
The four layers of cost
A website quote is always the sum of four things, even when they appear bundled into a single line.
Strategy and structure. Deciding what the site has to do, for whom, and in what order. Information architecture, page map, what goes on the homepage and what stays off it. It's the invisible layer, and it's the one that most determines whether the site works at all.
Content. Text, photography, video, illustration. This is the layer that most often vanishes from quotes, because everyone assumes the client is handling it.
Design. Drawing each page type, defining the visual system, resolving the in-between states — forms filled in badly, empty lists, error pages.
Build. Turning the design into code, connecting it to a content management system, testing across devices, making sure performance and accessibility hold up.
When two proposals differ by a factor of five, it's almost always because one includes all four layers and the other includes one and a half.
The lowest quote
The small number isn't necessarily a bad proposal. It's an honest proposal about a small scope — as long as it says so.
Typically, what's inside is: a theme or template adapted, your brand's colours and logo applied, your pages assembled with the content you supply, and delivery. That's real work and it has a legitimate place. For a small company that mostly needs to exist online with some credibility, it can be exactly the right call.
What isn't inside, almost always:
- Writing. You'll be writing it. If you've never written text for the web before, this is more work than it looks, and it's where most projects get stuck for months.
- Photography. Either you use what you have, or you buy stock, or the site keeps the template's generic images.
- Structural work. The architecture is the template's. If your business doesn't fit a structure designed for someone else's business, you force the fit.
- Fine tuning. The template works beautifully with its demo content. With yours — longer headings, a team of seventeen people, a service that needs explaining — it rarely works as well, and stretching it until it does is work that wasn't quoted.
- Support after launch. It ships and it's over.
The real risk with low proposals isn't the quality of what they deliver. It's what happens when the project grows halfway through — and it always grows, because the brief was incomplete. At that point you either pay for extras one by one, which added together overtake the mid-range proposal, or you cut scope and end up with a site that doesn't solve the problem.
The highest quote
On the other side, the big number isn't automatically expensive either. It's expensive if it's selling you things you don't need. It isn't expensive if it's including work you'd have to do anyway.
What usually justifies the difference:
- Content produced from scratch. Someone interviewing you, understanding the business and writing the pages. This is among the most expensive items and among the ones that change the result most — because what distinguishes your site from your competitor's is almost always the text, not the layout.
- Original design. Pages drawn for your content, with a system of their own. Instead of adapting the brand to a template, the template is built out of the brand.
- A photo shoot. Real people, real spaces, real work. It's the most visible difference between a site that looks like a specific company and a site that looks like a category of company.
- Serious technical work. Performance measured and optimised, accessibility, structured data for search, old URLs migrated with redirects so you don't lose the positions you had.
- More rounds and more people. A project involving art direction, copy and development brings in several heads and several revisions. That costs money.
There's also the stuff that inflates without adding: elaborate animations your audience doesn't value, a complex content management system for a site that changes twice a year, dozens of components designed for hypothetical pages you'll never create. It's worth asking, line by line, what each item actually solves.
The three factors that move the price most
Beyond the layers, there are specific variables that carry disproportionate weight.
The number of page templates, not the number of pages. A site with forty identical service pages is cheaper than a site with eight pages that are all different. When you ask for a quote, count types, not URLs. It's the question that separates someone who knows how to quote from someone who's guessing.
Who produces the content. This is the single biggest source of variation. If you write and you have photographs, the project is substantially cheaper. If you don't write and you don't have them, somebody has to — and the cost exists whether or not it appears on the proposal.
What happens after launch. A website isn't a finished object. It needs security updates, fixes, and someone who answers when something breaks. Proposals that include maintenance look more expensive in the first year and are cheaper by the third.
The cost nobody quotes for
There's one item that never appears on any proposal and is among the largest: your time.
A website demands decisions from you. Reviewing structure, approving design, writing or validating text, gathering scattered information, coordinating colleagues who have to contribute content. Projects rarely run late because of the agency and very often because the client doesn't have hours available for the decisions only they can make.
This has consequences for how you read the quotes. A more expensive proposal that includes content production is, in practice, buying you weeks of your own time back. If those weeks don't exist in your diary, the cheap proposal isn't cheap: it's a project that sits half-finished for a year.
Questions that make proposals comparable
If you want proposals you can actually compare, change the brief before you send it. Say explicitly:
- How many different page types you expect, and what they are.
- Who writes the text. If you don't know, say you don't know and ask for the figure with and without.
- What photography exists, and in what condition.
- Whether there's a brand guide, or whether the site defines the visual language.
- How many languages.
- Who will manage the site afterwards, and how technically comfortable they are.
- What date matters, and why.
- What order of magnitude you have available.
That last one frightens a lot of people, who worry the figure will rise to meet whatever number they name. The effect is usually the opposite. With no reference point, a studio quotes for the project it imagines, and that's nearly always bigger than the one you need. With a reference point, it quotes for the best possible project within what exists — and tells you what's being left out.
How to choose without choosing the number
Read each proposal trying to answer three things. What does this proposal assume I'll be doing? What happens if I want to change something halfway through? And what has this person understood about my business that the others didn't?
The third is the most revealing. A proposal that restates your problem more clearly than you wrote it is demonstrating the work before doing it. A proposal that repeats your brief in different words and attaches a price is selling execution — which may be plenty, if you know exactly what you want.
The lowest quote and the highest quote can both be right. They're simply answering different questions. What you have to work out is which of those questions is yours.