Two or three proposals are open on your desk and they cannot be compared. One runs to two pages, another to fourteen. One mentions hosting, one does not mention it at all, and a third includes it for a year without saying what happens in year two. Every one of them says content is included, and not one says who is writing it. The prices differ by a multiple, and there is no way to tell from the documents whether that gap is quality, scope, or somebody quietly leaving out the parts that get expensive later. This is not a failure of attention on your part. You are being asked to compare three descriptions of three different jobs and pick the cheapest one. The way out is not to read them harder. It is to make all three describe the same job first, which takes an evening and turns the comparison into arithmetic.
Key Takeaways
Normalise first. Write down one list of everything the project needs, then mark each quote as includes it, excludes it, or does not say. The third column is where the trouble lives.
An exclusion you can see is a decision. An assumption nobody wrote down becomes a change request later, and change requests are priced after you have already committed.
Who holds the domain registration, whose accounts the site lives in, and whether the design and code transfer to you are separate questions from the price, and they decide what you keep if you leave.
Why can two quotes for the same website be impossible to compare?
Because the word website is doing an enormous amount of hidden work, and each provider has quietly filled in the gaps with their own assumptions.
When you asked three companies to quote, none of them had a specification. They each built one in their head from your conversation, and those three specifications are not the same. One assumed you would supply photographs. One assumed a template with your colours applied. One assumed a full design system, because that is what they always build. All three then priced their own assumption accurately.
That is why the total at the bottom is the least informative number in the document. It is the sum of a scope you cannot see. Two independent forums for web professionals were both running a thread with the same headline three weeks before this was written, asking what actually explains the huge range in website design quotes, which tells you the people writing the quotes find the spread as hard to explain as the people receiving them.
So, the job is not to judge the prices. It is to reconstruct the three hidden specifications until they match, and then look at what is left.
What is the 3 quote rule, and is it a real rule?
It is a purchasing convention, not a rule, and nobody authoritative publishes it as one. It is worth knowing exactly what it does and does not do for you.
The useful part is real. Getting three written estimates rather than one is standard consumer advice for any substantial job, and the Federal Trade Commission's guidance on hiring contractors makes the same point plainly: get estimates in writing from more than one provider, and be wary of pressure to decide immediately. Three is not magic. Three is simply the smallest number that lets you see whether an unusual quote is unusual, because with two you cannot tell which one is the outlier.
The part that gets misused is the assumption that three prices are comparable because there are three of them. They are not. Three quotes for three different jobs produce a confident-looking decision built on nothing, and the cheapest of the three will usually be cheapest because it is the smallest, not because it is the best value.
One more thing the convention gets right, and it is worth stealing: give every provider the same written brief. Not a specification, just a page saying what the site must do, who writes the content, when you need it, and what you expect to own at the end. The quotes that come back will already be closer to comparable, and the one that ignores your brief has told you something useful for free.
How do you make two quotes describe the same job?
Build one list, then mark each quote against it. This is the whole method, and it works on any number of proposals.
Start by writing out every element the project needs, without looking at the quotes. Then read each proposal once, marking every element with one of three marks: included, explicitly excluded, or not mentioned. Resist the urge to interpret. If a proposal does not say who writes the content, it does not say, and that is the finding.
Normalising three proposals, in order
- 1
Write the list first, from your needs, not from their documents
If you build the list out of the quotes, you inherit their blind spots and will never notice what all three left out.
- 2
Mark each quote: in, out, or silent
Three columns. The silent column is the one to read again in the morning.
- 3
Turn every silent into a written question
One email, same questions to all three. Their answers are now part of the quote, which is the point.
- 4
Re-price only after the silences are filled
Some quotes will move. A number that moves a long way when you ask a simple question was never a quote, it was an opening position.
- 5
Compare like for like, then look at what is left over
When the scopes match, the remaining difference is genuinely rate, speed, and who is doing the work.
- 6
Then read the terms
Ownership, cancellation, what happens after launch. Same treatment, same three marks.
The email in step three is the best-spent half-hour in this process. You are not being difficult. You are converting assumptions into commitments, and a provider who resents being asked what is included has told you how the rest of the project will go.
Which line items are usually silent, and why do they matter?
The same handful of items go unmentioned in most proposals, and they are the ones that turn a cheaper quote into the more expensive project.
| Line item | What to look for | What silence usually means |
|---|---|---|
| Content | Who writes the words and sources the photographs, and by when | You are writing it, and the timeline assumes you already have |
| Number of unique page types | Distinct layouts, not total pages | The quote covers a handful of templates and the rest is a change request |
| Revisions | How many rounds, and what counts as one | One round, and the definition will be theirs |
| Integrations | Booking, payments, CRM, anything that talks to another system | Not included, and each one is its own small project |
| Migration | Keeping existing addresses working when the site moves | Not included, which is how a launch quietly loses you traffic |
| Accessibility | Conformance work as a named deliverable | Not included, and retrofitting it later costs more than building it in |
| Hosting and platform | What it costs after launch and whose account it is in | Bundled for a year, then a conversation you have not had yet |
| Training and handover | Whether you can change anything yourself afterwards | You will be raising a ticket to change a phone number |
Two of those deserve the most attention. Content is the item buyers most often assume is included and providers most often assume is not, and it delays more projects than every technical problem combined. Migration is the one nobody thinks about until enquiries drop after launch, because the pages that used to rank now return errors.
Who owns what when the work is finished?
This is a separate comparison from the price, and two quotes at the same number can leave you owning very different things.
Start with the domain name, because it is the simplest and most commonly mishandled. As the registrant you have specific rights and responsibilities, and ICANN publishes what those are for registrants along with a plain explanation of how registration works. The practical question is short: is the domain registered in your name and your organisation's account, or in your provider's? If it is theirs, moving away later is a negotiation rather than a decision.
Then the site itself. Ask whether the design files, the code and the content transfer to you, and get the answer in the agreement rather than in an email thread. Ask whose account the hosting lives in. Ask what you receive on the day the relationship ends, and what it would cost to leave. A provider confident in their work answers all of that comfortably, because retaining clients by holding their assets is not the same as retaining them by being good.
Finally, treat any promise about results the way you would treat any other advertising claim. A guarantee of rankings or leads inside a proposal is a claim being made to win a sale, and consumer protection guidance on classic warning signs of a scam is a reasonable calibration: urgency, pressure to sign today, and guarantees that cannot be explained are the same signals in any industry.
Could you normalise these three?
1. Quote A excludes content in writing. Quote B says nothing about content at all. Which is the better document?
2. You send the same three questions to all three providers and one quote rises sharply. What have you learned?
3. Two quotes match on price and scope. What separates them?
Pick an answer to begin.
Frequently Asked Questions About Comparing Web Design Quotes
Q: How many quotes should I get for a website?
Three is the usual convention and it is a sensible one, because two cannot tell you which is the outlier. It is not a rule and no authority publishes it as one. What matters far more than the count is that every provider quoted against the same written brief, since three quotes for three different jobs are worse than one quote you understand.
Q: How do I compare quotes that are wildly different lengths?
Ignore the documents' structure entirely and build your own list of what the project needs. Then mark each proposal as includes, excludes, or silent on every item. A fourteen-page proposal and a two-page one often normalise to nearly the same scope, and occasionally the short one is the more complete.
Q: What should I do about items a quote does not mention?
Ask, in writing, with the same questions to every provider. That single email converts assumptions into commitments and is the highest-value step in the whole process. Answers become part of what you are buying, and any quote that moves significantly in response was not a quote yet.
Q: Is the cheapest quote usually a bad sign?
Not by itself. It is usually the smallest scope, which can be exactly right if your needs are genuinely modest. It becomes a problem when the gap is made of things you will still need later, like content, migration or accessibility, because you will buy them anyway, at change-request prices.
Q: Should I tell providers what my budget is?
Yes, if you have a real one. It is not a weakness, it lets each provider propose the best version of the work at that level rather than guessing, and it makes the quotes far more comparable. What you should not do is share the other quotes, which turns a scoping exercise into an auction.
Q: What is a fair number of revision rounds?
There is no standard, which is why the definition matters more than the number. Ask what counts as one round and when the clock starts. Two clearly defined rounds beats unlimited revisions with no definition, because unlimited usually turns out to mean until the provider decides otherwise.
Q: How long should I expect a quote to be valid?
Most name a period, and if none does, ask. It matters less for the price than for what it tells you about how carefully the document was written. A proposal with no dates, no assumptions and no exclusions has not been thought about, whatever the number at the bottom says.
Wrapping Up
You are not comparing prices, you are comparing three different jobs that happen to share a name. Build your own list, mark each proposal as includes, excludes or silent, send one email that converts every silence into an answer, and only then look at what is left. The remaining difference is the real one, and it is usually much smaller than the totals suggested.
Do that once and you will never read a proposal the same way again. You will notice the blank rows before the numbers, which is the habit that separates a project that lands on budget from one that is still generating invoices in month five.
At Web Leveling we would rather be the quote you can read than the quote you have to trust. Our proposals say what is included, what is not, who writes the content and what you own at the end, and our web design work hands over the accounts and the files in your name, so leaving is always a decision rather than a negotiation. We work with small and medium businesses across the country and overseas, wherever they are. If you have two or three proposals in front of you right now, send them over and we will help you line them up, even if you go with somebody else.
Terms
The words that decide what a quote actually means
Tap a term to see what it means.
Scope. The written list of what is included. Everything expensive that happens later begins as something absent from here.
Exclusion. Something the quote says it does not cover. Genuinely good news, because it is visible and can be priced.
Assumption. Something the quote takes for granted, usually about you. The dangerous ones are the assumptions nobody wrote down.
Change request. Work added after the agreement, priced after you have committed. The single most common way a cheap quote becomes an expensive project.
Page type. A distinct layout rather than a distinct page. Twenty pages built from four layouts is a smaller job than eight pages built from eight.
Registrant. The person or organisation a domain name is registered to. If that is not you, moving your site later is a conversation rather than a decision.
Handover. What you actually receive at the end: accounts, files, documentation, and the ability to change something without raising a ticket.