Twelve questions to ask a web designer before you hire one.
Ask all twelve of everybody, write down what they say, and three quotes that looked impossible to compare will explain themselves inside an hour.
Twelve questions sort the field. Ask every designer or agency the same twelve, write down what they say, and the quotes that looked impossible to compare will explain themselves inside an hour.
The questions below are not trick questions and a good supplier will enjoy them. What you are testing is not whether somebody knows the right answer. It is whether they answer the question you asked, in a number or a sentence, without reaching for a brochure. Vagueness at the quoting stage is the single most reliable predictor of vagueness later, when it costs you money.
This page is built to be printed. Take it to the meeting.
Before you ask anybody anything
Have three things ready, because the quality of a quote depends almost entirely on the quality of what you hand over.
Know what you want the site to do. Book jobs, sell products, let people check you are real, take applications. One sentence is enough and most owners have never written it down.
Know roughly what you can spend. Refusing to say a number does not get you a better price. It gets you a quote padded for uncertainty, or three quotes for three different products that you then cannot compare.
Know what you already have. Who holds the domain, where the current site is hosted, whether anybody has the files. If you are unsure, that is worth twenty minutes on who actually owns your website before you go shopping.
The twelve questions
1. Who is writing the words on this website, and is that included?
The most important question on the list and the one most quotes dodge. Copy is where projects stall, and a quote that excludes it is cheaper for a reason you will discover in month three.
Good: a clear yes or no, with a named process either way, and a plan for how your knowledge gets into the copy.
Worrying: "you provide the content". Said quickly, then never mentioned again.
2. Can you show me three sites you have built that are still live today, and will one of those owners take my call?
Work in a portfolio can be unbuilt, unlaunched, or built by someone who has since left. Live sites with contactable owners cannot be faked.
Good: three URLs on the spot and an offer to introduce you.
Worrying: screenshots only, concept pieces, work they "collaborated on", or a reference who turns out to be a relative.
3. When this is paid for, whose name are the domain, the hosting and the code in?
You want all three in yours. The alternative is a business you do not fully control, which becomes obvious at the worst possible moment.
Good: everything in your name, them added as a manager or user.
Worrying: "we handle all that for you". That is an answer about convenience being offered in place of an answer about ownership.
4. What will this site cost me to keep running every year?
There is always a number. A custom static site can be close to the price of the domain. A platform build with licensed add-ons can run into four figures before anybody does any work.
Good: a line-by-line annual figure covering hosting, domain, licences and any subscription the site depends on.
Worrying: "basically nothing", or a monthly fee that bundles hosting, support and maintenance without saying what the parts cost.
5. After launch, who fixes something that breaks, and how quickly?
Things break. Forms stop arriving, a card expires, a plugin update takes the layout apart. What matters is whether there is a named person and a stated response time.
Good: a named contact, a channel, a response window, and the price of work outside it.
Worrying: "just email us any time". Pleasant, unenforceable, and usually accurate for about four months.
6. If I decide to move to somebody else in two years, what do I take with me?
Ask this in the first meeting, cheerfully. The reaction tells you more than the answer.
Good: you take the domain, the files or repository, the content and every account, and they will help with the handover.
Worrying: irritation. Or a build on a system only they can edit, with your content locked inside it.
7. Is this a fixed price, and when do I pay each part of it?
An hourly estimate transfers all the risk of their inefficiency onto you.
Good: a fixed quote in writing before work starts, split into phases, each invoiced as it is delivered.
Worrying: a range with no ceiling, a deposit of most of the total up front, or "we will know more once we get into it".
8. How many pages does this cover, and what does a page beyond that cost?
Scope creep is rarely malicious. It is usually two people who never agreed what the thing was.
Good: a numbered page list in the quote and a published price for an extra page.
Worrying: "as many as you need". That sentence will be reinterpreted later, and not in your favour.
9. What do you need from me, and by when?
Almost every delayed website is delayed by the client, usually on photos, product details or sign-off. A supplier who has done this before knows it and will tell you up front.
Good: a short list of inputs with dates attached, and a plain statement of what slips if they are late.
Worrying: nothing required from you at all. Either they are about to invent your business, or the request is coming later as an emergency.
10. Who will actually be doing the work?
The person selling is not always the person building. That is fine when it is disclosed and a problem when it is discovered.
Good: names, roles, where they are, and who you contact day to day.
Worrying: an evasive answer, or discovering mid-project that the work is subcontracted three time zones away and your feedback takes two days to land.
11. What is this going to be built on?
Not because one platform wins. Because the answer determines your annual cost, your speed, who else can work on it, and how hard it is to leave.
Good: a straight answer with the trade-offs stated, including the ones that do not suit them.
Worrying: a platform choice with no reasoning, or a proprietary system they built and only they maintain.
12. What are we measuring after launch to know whether it worked?
A website is a business expense and should be judged like one. If nobody defines success, the project gets judged on whether you personally like the colour.
Good: enquiries, calls, bookings or sales, tracked from day one, with a baseline taken before launch.
Worrying: impressions, page views, or a monthly report of numbers that move on their own.
Three answers that should end the meeting
Not warnings. Deal breakers.
- "We will own the site and you pay monthly to keep it up." This exists, it is marketed as affordable, and it means you are renting a business asset forever with nothing at the end. The total paid usually passes the cost of owning a custom build inside three years.
- "You will rank number one on Google." Nobody controls that, nobody can promise it, and the people who say it are describing a result they cannot produce.
- A quote you cannot get in writing. If the number only exists in conversation, there is no number.
What not to judge on
Some of the things buyers weigh heaviest are close to meaningless.
Agency size tells you about overheads, not about quality. Distance matters less than people expect, though somebody who knows your market already saves you explaining it. Awards are usually entered and paid for. The prettiest portfolio often belongs to whoever is best at photographing their work rather than whoever builds the best-performing sites.
And price alone decides nothing until you know what each price includes, which is why question one exists. Three quotes at wildly different numbers is normal and it is covered in what a website actually costs.
The printable list
- Who is writing the words, and is it included?
- Three live sites, and will an owner take my call?
- Whose name is the domain, hosting and code in?
- Annual running cost, itemized?
- Who fixes breakage after launch, and how fast?
- What do I take with me if I leave?
- Fixed price, paid when?
- How many pages, and what is an extra one?
- What do you need from me, and by when?
- Who is doing the work?
- What is it built on, and why that?
- What are we measuring afterwards?
Ask all twelve of everybody. Score each answer as clear, vague or avoided. In our experience the pattern across twelve answers predicts the project better than any single answer does, and you will know which supplier to pick before you finish adding it up.
We answer all twelve before quoting, and the timeline question is answered honestly in how long a website actually takes. If you want ours for your own job, send us the brief.
Who is writing the words, and is it included. Copy is where projects stall, and a quote that excludes it is cheaper for a reason you discover in month three, when the site is built and every page is still waiting on text from you.
No. Both belong in your name with the designer added as a manager. An answer along the lines of we handle all that for you offers convenience in place of ownership, and the difference becomes obvious at the worst possible moment.
It is the most useful five minutes of the whole process. Ask an owner three things: did the site launch on schedule, what did it cost against the quote, and what happened when they needed something changed six months after launch.
Prefer a fixed one-off price invoiced in phases. Monthly arrangements where the agency keeps ownership mean renting a business asset indefinitely with nothing at the end, and the total paid usually passes the cost of owning a custom build within three years.
A range with no ceiling, most of the money up front, no page list, no annual running cost, and any promise of a number one ranking. Each of those signals a scope that was never defined, and undefined scope is where overruns come from.