Ask five agencies to quote the same vague brief and you will get five numbers that share almost nothing except a currency symbol. That is not because agencies are inconsistent. It is because a brief that leaves scope open forces every agency to guess, and guesses get padded. A good website design brief is short, specific and honest about constraints. It takes an afternoon to write and it changes both the quotes you receive and the project you end up with.
A website design brief should cover your business, one measurable goal, your audience, the current site, required functionality, integrations, who writes content, budget range, timeline and who approves work. Agencies price uncertainty, so every unanswered question in the brief becomes a buffer in the quote.
What is a website design brief supposed to do?
A website design brief is a short document that tells an agency what you sell, who buys it, what the new site has to achieve, and what constraints exist. That is the whole job. It is not a spec, it is not a wireframe, and it is not a list of sites you like on Dribbble.
Most small business owners write the brief backwards. They start with pages and features because those feel concrete. The agency then quotes against a page count and everyone discovers in week five that the real requirement was a booking flow that talks to an existing scheduling system. That is a change order, and change orders are where budgets go to die.
Write the brief around outcomes and facts instead. Outcomes: what should happen more often after launch. Facts: what already exists, who has to approve, when it has to be live, what you can spend. An agency can design its way to an outcome. It cannot guess your approval chain.
Why do vague briefs produce inflated quotes?
Agencies price uncertainty. When a brief says "modern website for our services company, around 10 pages", the agency has to assume the worst plausible version of every unknown: content does not exist, there are four stakeholders with veto rights, the CRM integration is undocumented, and revisions will run long. Each assumption gets a buffer. Stack six buffers and the number doubles.
The reverse is also true and it is worse. Some agencies quote the best plausible version to win the deal, then recover margin later through change orders. A vague brief makes that easy to defend, because nothing in writing says the price included the thing you assumed it included.
In our own project experience, briefs that answer the content question, the integration question and the approval question up front tend to come back with tighter quotes and narrower ranges across the agencies you approach. That narrowing is the real signal. When three agencies quote within a reasonable band of each other, you are comparing like with like. When quotes range from $6,000 to $60,000, the brief is the variable, not the agencies.
Before sending anything, write one sentence: "This project is successful if ___ happens by ___." If you cannot finish that sentence, the brief is not ready and no quote you receive will mean much. Send that sentence to your team first and see whether they agree with it.
What goes in each section of the brief?
Ten sections is enough. Two pages is enough. Length is not the point, specificity is. Here is what belongs in each section and what a useful answer looks like.
| Section | What to write | Weak version |
|---|---|---|
| Business in one paragraph | What you sell, to whom, at roughly what price point, and how customers currently find you | "We are a leading provider of quality solutions" |
| Primary goal | One measurable outcome: more qualified enquiries, fewer support calls, online booking instead of phone | "Refresh our brand and look more professional" |
| Audience | Two or three real buyer types with the decision they are making and what would stop them | "Anyone who needs our services" |
| Current site | Platform, hosting, who built it, what works, what breaks, traffic sources, top pages | "It is old and we hate it" |
| Pages and functionality | Rough page list plus anything transactional: forms, quotes, payments, logins, calculators | A page count with no functional detail |
| Integrations | Named systems and whether they have an API: CRM, booking, email platform, accounting, inventory | Omitted entirely, then raised in week four |
| Content | Who writes the copy, who supplies photography, what exists today, what has to be created | "We will send content over" |
| Brand and design direction | Logo files, existing guidelines, three sites you like and one specific reason for each | Ten screenshots with no explanation |
| Budget and timeline | A range you can defend and a date with a reason attached to it | "Tell us what it costs" with a hard launch date |
| Decision process | Who approves, how many people review, how fast feedback comes back | Silence, then a surprise fourth reviewer |
The design direction row deserves a note. Reference sites are useful only with a reason attached. "I like this one" tells an agency nothing. "I like how this one puts the price range above the fold instead of hiding it" tells the designer what you actually value, which is honesty about pricing, not the typeface.
Which questions do clients most often fail to answer?
Four gaps show up again and again, and all four are expensive.
- Who writes the words? Content is the single most common cause of a stalled build. If the answer is "we will", set a date and name a person. If nobody has time, put copywriting in the scope and pay for it. A finished design sitting empty for two months costs more than the writing would have.
- What happens to the existing URLs? If your current site has pages that rank or receive traffic, someone has to map old URLs to new ones. Leave it out of the brief and it may be left out of the build, and you will feel it in organic traffic within weeks.
- Who signs off, and how quickly? Two reviewers with a three day turnaround is a normal project. Six reviewers with no deadline is a six month project regardless of how fast the agency builds.
- What has to keep working on day one? Enquiry forms routing to the right inbox, an existing booking link, a customer portal, tracking that finance relies on. List these explicitly. Anything unlisted is at risk during a migration.
There is a fifth, softer question: what have you already tried that did not work? If the last redesign failed, say why. Agencies that hear "we redesigned two years ago and enquiries did not move" will ask better questions than agencies handed a clean slate. Usually the answer is that the previous project changed the visual layer and nothing about the offer, the proof or the follow up.
How much budget detail should you share?
Share a range. The common fear is that stating a budget means the agency will spend all of it. In practice, withholding it means you receive quotes built on guesses, and you spend three weeks discovering that one agency was pricing a template build and another was pricing a custom design system.
A range does something more useful: it lets a good agency tell you your scope does not fit. That conversation is cheap in week one and brutal in week ten. If your budget genuinely cannot support the feature list, you want to hear it before design starts, so you can cut scope deliberately rather than have it cut for you.
Frame the range against the outcome. "We can spend in the mid five figures if this replaces our phone based booking" is a stronger statement than a bare number, because it tells the agency which parts of the build carry the return. That is the same logic behind good conversion optimization work: money follows the step that changes the result.
Send the same brief to every agency you shortlist, unchanged. Resist the urge to tailor it. Differences in the quotes then reflect differences in approach, not differences in what you told each of them, and the questions they ask back become a genuine comparison of how carefully they read.
How do you use the brief once quotes come in?
Read every quote back against the brief line by line and mark three things: what is included, what is excluded, and what is silent. Silence is the category that hurts. A quote that never mentions content migration has probably not priced content migration.
Then compare the questions each agency asked. An agency that came back with five sharp questions about your booking system has read the brief and is thinking about the hard part. An agency that returned a proposal within two hours and asked nothing has a template and is filling in your name. That signal is more predictive than the price, and it holds whether you are commissioning a brochure site, a Shopify build, or a full website design and development engagement.
Finally, insist that the accepted proposal restates the scope in its own words. If the agency can describe your project accurately without copying your sentences, they understand it. If the scope section is your brief pasted into their template, they have not processed it yet, and the first design presentation will show that.
What does the brief become after kickoff?
Keep it alive. The brief is the reference document you return to every time someone proposes something new, and someone always does. Halfway through, a stakeholder will want a video background or a team blog. The question is not whether it is a good idea in isolation. The question is whether it moves the outcome in section two.
Most scope creep is not malicious, it is drift. Ideas arrive one at a time, each small, each reasonable. Three months later the site launches late with a feature set nobody planned. A brief that states one primary goal gives you a way to say no that is not personal, because you are not rejecting the idea, you are noting it does not serve the stated goal and belongs in a phase two list.
Write that phase two list from day one. It is where good ideas go to wait instead of derailing the launch, and it becomes the natural starting point for whatever you do next, whether that is content work or paid acquisition once the site is actually converting.
Frequently asked questions
How long should a website design brief be?
Two pages is usually enough for a small business site. Specificity matters more than length. A short brief that names your integrations, your content owner, your approvers and a budget range is far more useful than ten pages of company history. If a section takes more than a paragraph, it is probably a requirements document rather than a brief, and that can follow later.
Should I include a budget in the brief?
Yes, as a range. Withholding it produces quotes built on guesses and makes comparison meaningless. A stated range also lets an agency tell you early that your feature list does not fit the money, which is a cheap conversation in week one and a painful one in week ten. Frame the range against the outcome you want rather than as a bare number.
What if I do not know what pages I need?
That is fine, and often better. Describe the outcomes and the actions you want visitors to take, then let the agency propose the structure. Client written page lists tend to mirror the internal org chart rather than how customers decide. What you should specify is anything transactional: forms, quotes, payments, logins, bookings. Those drive cost far more than the number of pages does.
Do I need wireframes before approaching an agency?
No. Wireframes drawn before anyone understands your audience tend to lock in assumptions that later need undoing, and they can make an agency feel obliged to build a layout they disagree with. Spend the effort on describing customers, sales process and constraints. Structure is part of what you are hiring a design team to work out with you.
Why do quotes for the same brief vary so much?
Usually because the brief left room for different readings of scope. One agency prices a template build with your content poured in. Another prices custom design, copywriting and a CRM integration. Both are answering what you wrote. Tightening the brief narrows the range, and when quotes cluster you are finally comparing agencies rather than comparing assumptions.
Should the brief say who the decision makers are?
Always. Approval speed drives timelines more than build speed does. Two reviewers with a three day turnaround is a normal project. Six reviewers and no feedback deadline stretches every stage. Agencies that suspect a crowded approval chain but cannot see it in the brief will price the risk anyway, so naming the process usually works in your favour.
The takeaway
The brief is the cheapest leverage you have on a web project. An afternoon spent naming your goal, your integrations, your content owner and your approvers will do more for the outcome than another week of comparing portfolios. Write it once, send the same version to everyone, and keep it open during the build. When someone proposes a new feature in week eight, the brief is what tells you whether to say yes.