Why Website Quotes Differ by 10x for the Same Brief

Send the same website brief to five suppliers and the quotes will not be close. That is not usually dishonesty or wildly different margins. It is that a website brief is almost always underspecified, and each supplier has silently filled the gaps with different assumptions.

Finding out which assumptions were made is more useful than negotiating on price, because the cheapest quote is frequently the one that assumed the least.

Content is the first hidden variable

A brief that says "about twelve pages" leaves open who writes them. Writing and editing twelve pages of copy is real work, often weeks of it, and it is regularly assumed away by both sides — the client expecting it included, the supplier assuming it arrives ready.

The same applies to images. Stock photography is cheap and looks it. Commissioned photography is neither. Diagrams, product shots and video each carry their own budget.

This single ambiguity explains a surprising share of quote variance, and it is also the most common cause of a project stalling after build, because the site is technically complete and there is nothing to put in it.

Design from scratch versus design from a system

"Custom design" covers a range from adapting an existing theme to originating a visual identity, and the cost difference between those endpoints is large.

A supplier quoting low is often planning to configure a template, which is a legitimate and sometimes correct choice. A supplier quoting high may be planning original design work including a component system, states, responsive behaviour and accessibility. Both can be right for different situations; they are not the same purchase.

Worth asking directly whether the price includes original design, how many rounds of revision, and what happens when you want something the underlying system does not do.

Integrations are where the estimate actually lives

A brochure site is a well-understood, fairly predictable build. A site that connects to a CRM, a payment provider, a booking system, a stock feed and an email platform is a different project wearing the same description.

Each integration carries its own uncertainty, and the variance is enormous: a modern documented API is a day, while a legacy system with no sandbox and undocumented behaviour can be weeks, most of it spent discovering how it actually responds rather than writing code.

If integrations are in scope, they should be listed individually in the quote. A single line saying "integrations" is a number someone guessed.

Who edits it afterwards

A site your marketing team can change without a developer costs more to build than one where every change is a code change. Content modelling, editor interfaces, preview, and guardrails that prevent someone accidentally breaking a layout are all work.

Skipping it is a defensible decision if the site genuinely will not change. It is a poor one if it will, because the saving converts into a support cost and a queue.

This is worth deciding on the evidence of how often you actually changed your last site, rather than on the intention to keep it fresh.

Performance and accessibility are scope, not qualities

Both are frequently assumed and rarely specified. A site can be built to load quickly on a mid-range phone over a poor connection, or it can be built to look right on the developer's laptop. The first involves image strategy, script discipline and measurement; the second does not.

Accessibility likewise. Building to WCAG 2.2 AA and testing with a screen reader is a real activity. If it is not in the brief, assume it is not in the quote, and if you have a legal obligation, that is a problem you will discover at the worst time.

What ongoing costs look like

The build is a one-off; the site is not. Hosting, domains, certificates, third-party subscriptions, dependency updates, security patches and content changes all continue.

Dependency maintenance is the one most often ignored. A site left unpatched for two years is a security problem and an expensive upgrade, because the gap has grown large enough that updating is no longer routine. Budgeting a modest amount continuously is considerably cheaper than a rescue project later.

Fixed price and time-and-materials price different risks

A fixed price includes a contingency for everything the supplier cannot see, and on an underspecified brief that contingency is large. You are paying for their uncertainty, and if the project turns out to be simple you do not get the difference back.

Time and materials removes the contingency and moves the risk to you. It works when you trust the supplier and can supervise scope. It works badly when neither is true, because there is no natural brake.

The middle position that usually serves clients best is a paid discovery phase — a week or two producing a specification detailed enough to quote against — followed by a fixed price on that specification. You can take the specification elsewhere, which keeps the quote honest.

Ask what happens when you disagree with a design

Revision rounds are where budget disputes concentrate. A quote including "two rounds of revision" is meaningless without a definition of what a round is: a batch of comments, a wholesale direction change, or anything the client says after seeing it.

Worth agreeing up front what constitutes a change of scope versus a refinement, and what happens if the direction is wrong after the first presentation. Suppliers who have been doing this a while will have a clear answer, and the clarity of that answer tells you something about how the project will run.

How to get comparable quotes

Specify who writes the copy and provides images. State whether design is original or template-based. List each integration by name and system. Say whether non-technical staff must be able to edit, and what. State performance and accessibility requirements explicitly. Ask for ongoing costs separately from build cost.

Do that and the quotes converge, because the suppliers are finally pricing the same thing. The remaining spread tells you something real about their approach rather than about their assumptions.