Building a small business website takes about three to five weeks of actual work. Most small business websites take three to six months to appear. That gap is the real subject of this article, because it is not made of work. It is made of waiting.
Ask five agencies how long a website takes and you will get five ranges, all of them wide, none of them explained. Nobody tells you what the range is made of, which parts are yours, or what would have to be true for a date to hold. So here is the anatomy, and then the conditions.
- Two numbers, not one. The work in a small business website is roughly fifteen to twenty-five working days. Elapsed time of three to six months is normal. The gap is waiting, not working.
- Four things stretch a build: queue time at the agency, approval latency at your end, scope that keeps moving, and content that arrives late. Only the first is the agency working slowly, and it is the rarest.
- A large agency quoting a range is being accurate rather than evasive. At eighty concurrent projects nobody can see far enough ahead to promise your date. The fix is structural, so asking harder does not produce one.
- Four conditions make any short timeline hold: scope fixed in writing on day one, review points scheduled with deadlines, content treated as a dependency, and a real cost to the agency when it slips.
- Longer is sometimes right. Large catalogues, custom software, brand work that does not exist yet, and big migrations all deserve a longer fixed timeline with its own gates rather than a short one that quietly slips.
The honest answer, in two numbers
A marketing website of up to eight pages, custom designed rather than dropped into a template, is roughly fifteen to twenty-five working days of skilled effort. Design, build, copy, technical setup, testing. That number moves with complexity. An eCommerce store with a catalogue and payment flows is more. A one-page site is less.
Elapsed time is the other number, and it is the one you feel. Three to six months is normal. Some builds run past a year. The work did not expand to fill that time. The work sat still inside it.
Once you separate those two numbers, the question changes. You stop asking how long the work takes and start asking what is going to happen to your project in between the days somebody touches it.
Where the months actually go
Four things stretch a build. Only one of them is the agency doing the work slowly, and it is the rarest of the four.
Notice how much of that column is yours. This is not an argument that delays are your fault. It is the reason vague timelines survive: when a project slips, responsibility is genuinely shared, so nobody is accountable and no date was ever binding. A timeline that does not name who owes what, and by when, is a forecast rather than a commitment.
Why you get a range instead of a date
A large agency is not being evasive when it says three to six months. It is being accurate. At eighty concurrent projects with shared designers and developers, no individual can see far enough ahead to promise your Tuesday. The range is honest reporting of a system that cannot produce a date.
Which means the fix is structural, not motivational. You do not get a firm date by asking harder. You get one from an agency small enough to see its whole queue, with a scope written down before anyone starts, and a cap on how many builds run at once.
There is a second reason, and it is less comfortable. A date with no penalty attached costs nothing to miss. If the only consequence of launching in month five instead of month three is an apologetic email, then the date was never a constraint on anybody's behaviour. It was a hope, stated confidently at the start of a sales conversation.
What has to be true for thirty days to hold
We launch in thirty days or the build month is free, and the whole day-by-day process is published on the 30-day launch page. Rather than repeat the table here, it is worth pulling out the four conditions that make any short timeline hold, whoever is running it.
- Scope is fixed in writing on day one. Sitemap, page count, what is included and what is not. Not to make the project rigid, but so that finishing means something. An unfixed scope has no finish line to hit, and a deadline without a finish line is decoration.
- Your review points are scheduled, with deadlines. Ours are day 10 and day 22, and we need your notes within two business days at each one. Scheduling them in advance is what converts approval latency from an invisible risk into a date in your calendar you can plan around.
- Content is treated as a dependency, not an afterthought. Copy is written as part of the build, not requested from you in week three. If photography is needed, it is booked in week one, because a shoot that gets arranged in week four is the shoot that costs you a month.
- Somebody pays when it slips. If the delay is on the agency side, the build month is free. If the agency is waiting on you, the clock pauses and restarts when the content arrives. Both halves matter. A one-sided guarantee gets quietly abandoned the first time a client is slow, and everyone learns the date was soft after all.
None of these require a special agency. They require an agency willing to write down what it owes you and attach a cost to missing it. That most will not is a choice about risk, not a limit of the craft.
Five questions to ask before you sign
Whoever you hire, these five separate a real date from a hopeful one. Ask them in the first call, and pay attention to which ones produce a pause.
- How many builds are running at the same time as mine? This is the queue-time question. A number, however large, is a better answer than a change of subject.
- What happens if you miss the date? If the answer contains no consequence, the date is a forecast. That can be acceptable, as long as you both know that is what you bought.
- On which specific days do you need something back from me? A good answer is two or three named review points with response windows. A vague one means approval latency is going to be discovered rather than planned.
- Who is writing the copy, and when? The most common reason a build stalls is that nobody agreed the answer to this until the site was otherwise finished.
- What is explicitly not included? An agency that can name the exclusions has a fixed scope. One that says everything you need has not defined the project yet, and the extras will arrive later as change requests.
When a longer build is the right answer
Thirty days is not always correct, and pretending otherwise would be the same trick from the other direction.
A large eCommerce catalogue with complex variants and stock integration needs longer. So does a site that depends on custom software, a booking engine wired into an existing system, or a database of properties or listings. If your brand does not exist yet, the positioning work should happen before the design work rather than inside it, and that is a separate project with its own timeline.
Migrating a large site with years of accumulated URLs is its own discipline, and rushing the redirect mapping is how businesses lose rankings they spent years earning. In each of these cases the right response is a longer fixed timeline with its own gates, not a short one that quietly slips.
What should not stretch a build is a straightforward marketing website for a business that already knows what it does. That is the project this article is about, and three to six months for it is a queue problem being presented as a craft problem.
The fifteen to twenty-five working day figure is our own estimate of skilled effort on a custom marketing site of up to eight pages, drawn from our build history rather than from a survey. The three to six month elapsed range describes what is commonly quoted across the agency market and what ranks for this query today; it is not a measured industry average, and no published study is claimed for it. The four causes of delay are our editorial account of why the two numbers diverge, not a ranked measurement. The day 10 and day 22 review gates, the two business day response window, the five lines of contract terms and the clock-pause rule are our published terms, set out in full on the 30-day launch page. The $4,900 build price and its itemisation are our published prices, listed on the pricing page.
The timeline and the price are the same question
An agency that cannot tell you what a website costs, and 79% of the top fifty will not, usually cannot tell you when it will be finished either, and for the same underlying reason: neither has been defined. Scope drives both. Once the sitemap, page count and inclusions are written down, the price stops being a negotiation about what you look able to pay and the date stops being a guess.
Ours is $4,900 for a custom build of up to eight pages, itemised line by line on the pricing page, with the thirty-day guarantee included rather than sold as an upgrade. If you want the longer version of the cost side, we wrote what a website actually costs as its own piece.
The practical test for any quote you are holding: does it name a date, name what is included, and name what happens if the date is missed? If all three are present, you have a commitment. If any is missing, you have an estimate, and estimates are where months go to hide.