Dubai · Web Design
Website redesign services in Dubai — a rebuild that keeps what already ranks.
Your current site is audited page by page before anything is drawn: what ranks, what converts, what nobody has opened in a year. You finish with a new site, a full redirect map, and every URL worth keeping still resolving where it did.
- Redirects mapped before launch
- Weekly builds on a staging URL
- Rankings monitored through launch
What's included
Six parts, and four of them are migration.
The design and the build are the visible half. The other half is making sure a site that already earns its traffic still earns it on the Monday after launch.
A page-by-page audit of the live site
A crawl of every indexed URL, read alongside whatever analytics and Search Console history exists: what gets traffic, which queries it arrives on, what converts, and what has had no visitor in a year. The output is a decision beside each page — keep, merge or retire.
A redirect map, one row per URL
Every old address assigned a destination on the new site, written down and reviewed while it is still cheap to argue about. A wildcard rule that sweeps everything to the homepage is the most common way a redesign quietly loses its rankings, and it is not used here.
Content migrated with its metadata
Titles, meta descriptions, canonical tags, publication dates and image alt text move with the body copy rather than being retyped and quietly dropped. Where two pages merge, the surviving one is written to cover what both of them answered.
Structure re-decided against evidence
The new sitemap is drawn against the audit rather than against the old navigation, so pages nobody used stop taking up room and pages that quietly carry the traffic get more of it. This is where a redesign either fixes the underlying problem or repeats it in a nicer typeface.
A rebuild with a speed budget
The new site is built against a performance budget agreed before the first line of code rather than measured hopefully afterwards. Sites due for replacement are usually slow because of what accumulated on them, and a rebuild is the one moment you are not obliged to carry any of it across.
The weeks after launch, watched
Crawl errors, coverage and index status read in Search Console after cutover rather than assumed. Redirect chains, missed pages and a sitemap still listing the old URLs only appear once the site is live — and all of them are cheap to fix in the first week.
Who it's for
Three reasons a rebuild is the right call.
If the pages perform and only one step in the journey leaks, replacing the whole site is the expensive answer. Tell us that early and we will point you at the cheaper one.
A site that still earns, and still embarrasses
The traffic is fine. The design is eight years old, the mobile layout was an afterthought, and you have stopped sending the link to prospects. Nothing about the search performance justifies starting over — which is exactly why the URLs and the page structure have to survive the rebuild rather than be redrawn for convenience.
You describe the site apologetically before anyone opens it.
A platform that became the constraint
Every change needs a developer, the page builder has accumulated a decade of plugins, and the last update broke something unrelated. The design is not really the complaint; the thing it was built in is, and restyling a site does not change what it is made of.
Your last content edit went through an agency, not your team.
A rebrand, a merger or a second language
The company the site describes is not the company any more — a new name, a wider offer, or Arabic that the original build left no room for. Those are structural changes: URL structure, content model and navigation all move together, which is a rebuild whatever it gets called.
You need pages the current site has nowhere to put.
How we deliver it
Five stages, and the first two are not design.
The order matters more here than on a new build. Everything that protects the site you already have is settled before anything visual is drawn.
01
Audit and inventory
A full crawl of the live site, read against whatever analytics history exists. Ends with a table of every URL carrying a keep, merge or retire decision — signed off by you before anyone opens a design tool.
02
Structure and redirect map
The new sitemap drawn against that inventory, then every old URL assigned its destination. Both are agreed before design starts, because a structure decided later is a redirect map rewritten later.
03
Design
Wireframes, then interface design in Figma as components rather than screens. On a typical six-week project this runs roughly week two to week four, working from an audit that has already settled what each page has to do.
04
Build beside the live site
The new site goes up on a staging URL you can visit weekly while the old one keeps running and keeps being indexed. Content is migrated here, with its metadata, rather than after the switch.
05
Cutover and the weeks after
Redirects deployed and tested against the inventory before DNS changes, then the sitemap resubmitted and Search Console read for crawl errors and coverage. A missed URL costs minutes at this stage; found six months later it is a project.
Stack
Boring technology, chosen so the next developer recognises it.
Nothing here is exotic. That is the point: an unusual stack is a lock-in mechanism dressed up as innovation.
- Front-end
- Next.jsReactTypeScriptTailwind CSSAstro
- CMS & commerce
- WordPressShopifySanityPayloadWooCommerce
- Infrastructure
- VercelAWSCloudflareGitHub ActionsSentry
Deliverables
What lands in your hands at the end.
Two of these are documents you can check against the live site yourself, which is the only way to know a migration was actually done rather than described.
- A URL inventory of the old site — every page with its traffic, entry queries and conversions, and a keep, merge or retire decision beside it
- The redirect map, as a reviewable table and as the rules actually deployed: one row per old URL, and no wildcard sweep to the homepage
- A full backup of the old site and its database, taken before anything is switched over
- The new site live, with the repository, the Figma file and the hosting in your accounts rather than ours
- Content migrated with its metadata intact — titles, descriptions, canonicals and publication dates, not only the body copy
- Analytics and Search Console reconnected, with events and goals re-tested on the new templates rather than assumed to have survived
- A post-launch crawl checked against the inventory, listing every URL that resolved and every one that did not
- Editor training on the new CMS, recorded
The redirect map is the part most redesigns skip, and it is the part that decides whether Monday's traffic arrives. It is deployed and tested before DNS changes — not written up afterwards as an explanation.
Pricing
What a redesign costs.
AED 16,000
from
A redesign carries the audit, the redirect map and the content migration on top of whatever the new site itself costs to build — which is why the page count of your current site will not tell you the tier on its own. AED 16,000 is the Launch tier, five to eight pages. A larger page set, a custom content structure or Arabic sits in the Growth tier from AED 48,000, A store sits in that same Growth tier; portals, larger catalogues and stock synced to an ERP start at AED 95,000. Prices exclude 5% VAT.
What changes the priceWhat that covers
- A page-by-page audit of the live site
- URL preservation and a full redirect map
- Content migrated with its metadata
- Custom design, never a bought theme
- A performance budget agreed before the build
- Search Console watched through cutover
Real estate
The URL count moves every week. Listings are added, sold, withdrawn and relisted, and a redesign has to decide what a sold property's page becomes rather than letting a few thousand of them 404 at once. The portal feed and the site also have to keep agreeing after the templates change.
Clinics and healthcare
Few pages, each carrying a lot. A single treatment page is often the entry point for the term it is named after, so merging three of them into one tidy services page trades three rankings for one. Practitioner profiles, insurer lists and the clinic's name, address and phone all have to stay consistent with what is published off the site.
A redesign is one of five services in the full web design service and it is the only one that begins with a site already live and already indexed. If nothing has been designed yet, or the structure itself is what failed, the work starts earlier than this page describes.
Questions
Asked before a rebuild is signed off.
Not if the URLs are preserved and every changed one is redirected — which is the part of a redesign that gets skipped most often. Rankings attach to addresses, so a page that moves without a 301 loses what it earned. We map every old URL to a destination before design starts, deploy the rules before DNS changes, then read Search Console for coverage errors afterwards. Some movement while Google recrawls is normal; a fall that does not recover is a redirect problem.
Each page gets one of three decisions — keep, merge or retire — made against traffic and search data rather than taste. Pages that earn visits are migrated with their titles, meta descriptions, canonicals and dates intact. Pages being merged have their material folded into the surviving version, and retired ones are redirected to the nearest useful page rather than to the homepage. You approve that table before anything is drawn.
Rebuilding a five-to-eight page site starts at AED 16,000, the Launch tier, with the audit and the redirect map inside that figure rather than quoted beside it. A larger page set, a custom content structure or Arabic moves it into the Growth tier from AED 48,000, A store sits in that same Growth tier; portals, larger catalogues and stock synced to an ERP start at AED 95,000. The migration half is scoped against the site you already have rather than the one being drawn, so the number of pages you have today is a poor guide to the tier on its own. Prices exclude 5% VAT.
It can stay, and often should. If your team publishes comfortably and the requirements are conventional, redesigning on the same platform removes a whole category of migration risk — the content stays where it is, and only the structure and the theme change. Moving is worth it when the platform is the actual complaint: a page builder nobody can maintain, a lapsed theme licence, or plugins doing work that belongs in code.
Yes. The rebuild happens on a staging URL alongside it, so the live site keeps trading and keeps being indexed until the day you switch. Staging is password-protected and excluded from search engines, which matters more than it sounds — an indexed staging copy competes with the site it is meant to replace. Cutover is a scheduled change, and the old site is backed up in full beforehand.
That depends on what made it slow, and the audit says which. Sites due for replacement are usually slow through accumulation — plugins, tag managers, four analytics scripts and an unoptimised hero image — and a rebuild is the one moment you are not obliged to carry any of it across. Every build ships against a performance budget agreed beforehand; if a change would break the budget, the change gets reworked.
They are rebuilt and re-tested rather than assumed to survive. Conversion events, form submissions and tag manager triggers are all bound to page structures that a redesign changes, so each is checked on the new templates before cutover and again after. Search Console keeps its history because the property belongs to the domain rather than to the site — which is why annotating the launch date in analytics is worth the two minutes.
It is workable, as long as you control the domain. Hosting and CMS access make the migration cleaner, but a site can be inventoried from a public crawl and its content pulled from the live pages. The domain is the thing that cannot be recreated: without registrar or DNS control, the new site cannot go live on your address and the redirects have nowhere to run. Recovering that access is task one.
Ready to build something that lasts?
Tell us what you need. You will get a scoped response with an indicative price — from an engineer, not a salesperson.