Weeks, not months. On a modern stack the build is the fast part. What stretches a project is unwritten copy, review cycles measured in days of waiting rather than hours of work, and requirements discovered after sign-off. Fix the scope, write the words first, and a business site is a one to two week job.

This is not an argument that everyone else is slow. It is an argument about where the calendar actually goes, because once you can see that, you can compress it. Almost none of it is engineering.

The build is not where the time goes

Assembling a ten page marketing site is no longer the hard part of a website project. Component libraries, hosting, forms, analytics and structured data are all commodity now. Given a settled design and finished copy, the assembly is days.

That is worth saying plainly because it reframes every timeline conversation you will have. If a proposal says four months, the four months are not being spent on the thing the proposal is nominally for. They are being spent on deciding what to say, waiting for people to look at it, and absorbing requirements that arrived late. Those are real costs. They are just not build costs, and they respond to completely different treatment.

Where the weeks actually go

Three things, in order of how much damage they do.

Copy that does not exist yet. This is the big one. Design cannot be finished around placeholder text, because layout follows length and hierarchy follows argument. A headline of six words and a headline of eighteen are different designs. Every project where the copy is arriving next week stalls in the same place, and it stalls after the design has been approved, which is the most expensive moment to stall.

Review rounds priced in calendar rather than hours. A round of feedback costs someone an hour of attention and the project a week of elapsed time, because it waits on a person who has a business to run. Four rounds is a month, and the work inside that month is four hours. This is why uncapped revision cycles are the single most reliable way to turn a short project into a long one.

Requirements that arrive after sign-off. A CRM to connect. A booking system. A second language. A permit number that has to appear on every listing. Each of these is small on its own and each one reopens decisions that were already closed.

The pattern is the same in all three: the expensive thing is not the work, it is the re-deciding. A scope that is genuinely fixed before anyone opens a design tool removes most of it.

The UAE requirements that surface after sign-off

Two are worth knowing before you brief anyone, because both are cheap on day one and expensive in week six.

If you advertise property, your listing template is a compliance surface. The Dubai Land Department requires a Trakheesi permit number in real estate advertisements, and since 24 April 2023 it has required a Madmoun QR code on advertisements so a buyer can verify the listing, the advertising company and the property details. DLD enforces this: in one announced action it fined ten companies and warned thirty more for missing permits, misused permit numbers and expired permits, at AED 50,000 per fine and escalating to licence cancellation for repeat violations. A brokerage site therefore needs a listing component that carries a permit field and a QR code as standard. Designed in, that is an afternoon. Retrofitted across a finished site, it is a rebuild of the one template everything else depends on.

If you have a contact form, you are processing personal data. Federal Decree-Law No. 45 of 2021, in force since 2 January 2022, prohibits processing personal data without the consent of the data subject, subject to a list of exceptions, and gives people rights to correct their data and to restrict or stop its processing. The interesting part is the gap: the implementing regulations have still not been issued and the Emirates Data Office is not yet fully operational, so enforcement has been light and most organisations are driving their own compliance. The obligation exists while the detail and the enforcement do not. That is a reason to put a plain consent line and a short privacy page in the first brief rather than a reason to skip it, because the cost of doing it on day one is close to zero.

What actually decides whether the site earns anything

Two things, and both are settled in the first week: what it says, and how fast it is.

Speed is the one people underrate, because it feels like a technical detail rather than a commercial one. Google's Core Web Vitals set the bar at a Largest Contentful Paint within 2.5 seconds, an Interaction to Next Paint of 200 milliseconds or less, and a Cumulative Layout Shift of 0.1 or less. The detail that catches people out is how they are assessed: at the 75th percentile of real page loads, segmented across mobile and desktop. So the one-off test you ran on your laptop is not the number in use. Field data from your actual visitors, on their actual phones, is.

What that is worth commercially is unusually well measured. Deloitte Digital and Google studied 37 European and American brand sites across more than 30 million sessions, monitoring mobile load times for 30 days, and published the results as Milliseconds Make Millions. A 0.1 second improvement in mobile load time produced a 21.6 percent increase in progression to form submission on lead generation sites and a 7 percent increase in page views. Retail saw conversion up 8.4 percent and average order value up 9.2 percent, travel saw conversion up 10.1 percent, and in luxury, progression to the Contact Us page rose 20.6 percent.

Two caveats worth stating, because the number gets quoted carelessly. The data is from the end of 2019, and the sample is European and American rather than Gulf. Treat the direction as the finding rather than the decimal place.

Even discounted, the implication is hard to argue with. If your site's entire job is to turn a visitor into an enquiry, a tenth of a second is worth more than another round of revisions on a colour. And a tenth of a second is a decision made in week one about how the thing is built, not something you can add in month three.

What this means in practice

If you want a website in weeks rather than months, the work that compresses the timeline happens before anyone opens a design tool.

Write one page before you brief anybody. Who this is for, what you do, why you rather than the obvious alternative, and three pieces of proof. If you cannot write that page, no agency can write it for you in month three. They will simply take longer to discover that it was missing, and you will pay for the discovery. This single page is also the thing that makes the copy stage fast, because it settles the argument the site has to make.

Fix the scope in writing, including what is excluded. An exclusions list is more useful than a features list. Ecommerce, custom integrations, ongoing content, translation of copy rather than of the interface: name what is out, so that adding it later is a visible decision rather than a silent slip.

Name one decision maker and one content deadline. Most overruns trace to a missing person rather than a missing developer. Cap the review rounds, put them in the calendar in advance, and treat a missed content deadline as the schedule risk it actually is.

Ask about ownership before you start, not at handover. Source code, hosting and the domain registered in your own name. This is easy to agree at the beginning and awkward to raise at the end, which is exactly why it should be in the first conversation.

Do all of that and a business website is a one to two week project, which is how we run ours. If a proposal in front of you quotes months, the useful question is not whether the price is fair. It is which of these four things the months are being spent on, and whether that is engineering or scheduling. The answer tells you what you are actually buying.

If you want a view on a proposal you have in front of you, or on what your current site is costing you in enquiries, tell us what you sell and who you are selling to.

Sources

UAE regulations change frequently, and how any specific rule applies to your business depends on facts an article cannot see. Verify the current position before you decide. This is commercial guidance, not legal advice.