What do technical SEO services in Dubai include?
Technical SEO is the work that makes sure search engines can find, fetch, render, understand and index the pages you want found, quickly and without confusion. Content and links decide how well a page can rank; technical SEO decides whether the page is in the race at all.
On a small brochure site, that is a short checklist. On a large UAE site, technical SEO services in Dubai become an engineering discipline, because the problems multiply with URL count. A property portal might generate listing pages, area pages, building pages, agent pages and filter combinations, in two languages, with listings expiring daily. A marketplace adds sellers, variants and sort orders. Each of these can produce thousands of low-value URLs competing for Google's attention with the pages that actually earn leads.
The work breaks into six areas, and a good audit covers all of them before recommending anything.
- Crawling: robots.txt, internal links, parameters, crawl budget and server capacity
- Indexing: status codes, canonicals, noindex rules, duplicates and sitemaps
- Rendering: whether content and links exist after JavaScript runs
- Language and region: hreflang, URL structure and language-specific canonicals
- Performance: Core Web Vitals from real users on mobile
- Understanding: structured data, titles, headings and internal anchor text
When does a UAE site need a technical SEO audit rather than more content?
Commission a technical audit when you publish good pages that do not get indexed, when traffic falls after a release, or when Search Console shows a growing pile of excluded URLs. Writing more content onto a site with a crawling or indexing problem only adds to the backlog.
The clearest signal is in the Page indexing report. A large and growing count under “Discovered – currently not indexed” on a listings site often means Google knows the URLs exist but is not getting round to fetching them, which points at crawl efficiency or perceived quality. “Duplicate without user-selected canonical” in bulk usually means filters or language versions are colliding. “Soft 404” counts on a property site often trace back to expired listings that still return a normal page.
Other signs are commercial rather than technical: new projects or products taking weeks to appear in search, Arabic pages ranking for English queries, a sharp traffic drop after moving to a new framework, or mobile rankings trailing desktop. Each has a technical explanation that an audit can find in days.
- New listings take more than a few days to be indexed
- Excluded URLs in Search Console outnumber indexed ones
- The wrong language version appears in results
- Traffic dropped after a redesign, migration or framework change
Does crawl budget matter for UAE property portals and marketplaces?
Usually yes, because these are the sites Google's guidance is written for. Google's crawl budget documentation says it is aimed at roughly sites with more than a million unique pages changing about weekly, sites with more than ten thousand unique pages changing daily, and sites with a large share of URLs stuck as “Discovered – currently not indexed”. A busy UAE listings site often fits the second group.
Crawl budget is the combination of how much Googlebot can crawl without straining your server and how much it wants to crawl based on the value it sees. You influence both. Faster, more reliable server responses raise capacity. Removing duplicate and junk URLs raises the share of crawling that lands on pages that matter.
On portals, the waste usually comes from the same places: sort and filter parameters, map-view URLs, tracking parameters added to internal links, calendar and pagination traps, and old listings that never return a proper error. Server logs show exactly where Googlebot's requests go. Our technical SEO services in Dubai start there on any site with more than a few thousand URLs, because the log tells you what Google does, while a crawler only tells you what is possible.
What should happen to expired, sold or rented listings?
Return a real error for listings that are gone for good, and keep the page live only if it still helps a visitor. Google's crawl budget guidance recommends returning 404 or 410 for permanently removed pages rather than relying on noindex, because a 404 is a strong signal not to crawl that URL again.
Property portals and classifieds face this at scale. A rented apartment in Dubai Marina might have had a listing page for three weeks; once it is off the market, keeping it as a thin “this listing is no longer available” page with a 200 status produces soft 404s and wastes crawling. The better options depend on the case.
For a single expired listing, a 404 or 410 page that links to similar live listings in the same building or community keeps the visitor, while telling Google the URL is finished. For a building or project page that stays relevant after units sell, keep the page, update the content and link to the latest units. Avoid redirecting every expired listing to the home page, which Google may treat as a soft 404 anyway.
Whatever rule you choose, apply it in code by listing status, so the behaviour is consistent across tens of thousands of URLs and does not depend on editors remembering.
Listing gone, no replacement
Return 404 or 410 with links to similar live listings; remove it from the sitemap.
Project or building still relevant
Keep the page at 200, refresh the content and point to current units.
Listing moved to a new URL
A single 301 redirect to the new URL, updated in internal links and sitemaps.
Faceted search on marketplaces: which filter pages should Google see?
Let Google index only the filter combinations people actually search for, and keep the rest out of the crawl. Google's faceted navigation guidance warns that filter URLs cause overcrawling and slow the discovery of useful new pages.
A property search with filters for area, type, bedrooms, price band, furnishing and amenities can produce millions of combinations. A few of those match real queries, such as “2 bedroom apartments for rent in JVC” or “villas for sale in Arabian Ranches”. Those deserve static, well-linked landing pages with unique titles and content. The combination of “furnished, pet-friendly, 3 to 4 bathrooms, sorted by newest” does not.
Google lists several controls. If filtered pages do not need to be indexed, block their parameters in robots.txt or use URL fragments. If they must be crawled, use standard “&” separators, keep filter order consistent, and return 404 when a combination has no results. Canonical tags and nofollow on filter links are weaker, longer-term signals. The right mix depends on how your platform generates URLs, which is why we read the code before recommending rules.
How to set up hreflang for en-AE and ar-AE pages
Give every English page and its Arabic equivalent a set of hreflang annotations that list both versions and themselves, plus an x-default. Google's localised versions guide says the language code uses ISO 639-1, the optional region code uses ISO 3166-1 Alpha 2, and that if two pages do not point to each other the tags are ignored.
For a UAE-only site, the pair is usually “en-AE” and “ar-AE”, or simply “en” and “ar” if you also want to serve Arabic and English speakers elsewhere in the Gulf. You cannot use a country code alone. The annotations can sit in the HTML head, in HTTP headers, or in the XML sitemap; Google treats the methods as equivalent. On large sites, sitemaps are easier to keep accurate than template tags.
The most common errors we find on bilingual technical SEO audits are missing return links (the Arabic page lists the English one, but not the reverse), hreflang pointing at redirected or non-canonical URLs, Arabic pages whose canonical points to the English version, and language switchers that send users to the Arabic home page instead of the equivalent page. Each of these makes Google fall back on its own guess, which is how Arabic pages end up in English results.
Arabic URLs, RTL templates and duplicate language versions
Put each language in its own consistent path, such as /en/ and /ar/, give each version a self-referencing canonical, and make sure the Arabic page is genuinely translated rather than an English page with a mirrored layout.
Arabic slugs are a choice, not a requirement. Native Arabic slugs can match Arabic queries and read naturally in the address bar, but they appear percent-encoded when copied into some tools and emails. Transliterated or English slugs under /ar/ are simpler to manage. Either works if it is consistent and the hreflang map uses the exact final URLs.
Watch for partial translations. Many UAE sites translate the menu and headings but leave product specs, listing descriptions or FAQs in English. Google may see those Arabic URLs as near-duplicates of the English pages and pick one. Where translation is incomplete, it is often better to leave the page out of the Arabic version and its hreflang set until the copy exists. Translated copy comes from you or a professional translator; our role is the structure. Our Arabic website design page covers RTL layout and fonts in more detail.
Core Web Vitals on mobile: where large UAE sites usually fail
Most large UAE sites fail on Largest Contentful Paint and Interaction to Next Paint on mobile, not on layout shift. According to web.dev, good thresholds are LCP within 2.5 seconds, INP of 200 milliseconds or less and CLS of 0.1 or less, measured at the 75th percentile of page loads.
The causes repeat across portals and stores. Hero images and listing galleries are served at desktop size to phones. An embedded map loads on page start even though most visitors never scroll to it. Chat widgets, heatmap tools and several ad pixels compete for the main thread, which drags INP down on mid-range Android phones. Arabic web fonts load late and cause text to reflow. Consent banners inject themselves above the content.
Fixes are template-level, so one change can improve thousands of pages: responsive images with correct sizing, lazy-loaded maps behind a click, deferred third-party scripts, preloaded primary fonts with subsetting for the Arabic glyphs you use, and reserved space for banners. We measure with field data from the Chrome UX Report in Search Console, not only lab scores, because a fast result on a developer's laptop in Dubai says little about a commuter on mobile data.
Is Google indexing your JavaScript site properly?
Check the rendered HTML, not the source. Google's JavaScript SEO basics describe three phases, crawling, rendering and indexing, and note that pages can wait in the rendering queue for longer than a few seconds. Content and links that exist only after rendering are discovered later, and sometimes not at all.
UAE portals and marketplaces built as single-page apps in React, Vue or Angular often ship a near-empty HTML shell. Google may render it eventually, but internal links added by JavaScript are found late, and other crawlers, including many AI crawlers, may not run JavaScript at all. Google itself says server-side or pre-rendering is still a great idea because not all bots can run JavaScript.
The audit compares source HTML with rendered HTML for each major template using the URL Inspection tool and a headless browser. We look for titles, canonicals and hreflang set only by JavaScript, links built from click handlers instead of real anchor elements with an href, infinite scroll without paginated URLs, and content hidden behind tabs that never loads until clicked. Frameworks such as Next.js or Nuxt make server rendering straightforward once the templates are set up for it; older single-page apps may need pre-rendering as an interim step.
Structured data for listings and products: what Google actually uses
Mark up what Google supports and what visitors can see on the page. Google's search gallery includes Product, Local business, Organization, Breadcrumb, Vacation rental and others; it has no dedicated rich result for sale or long-term rental property listings at the time of writing.
For stores and marketplaces, product markup is the priority. Google's product documentation separates product snippets from merchant listing experiences and says that providing structured data plus a Merchant Center feed maximises eligibility. Price, availability, variants, shipping and return policy fields should match what the page shows in AED, including VAT treatment. Our ecommerce SEO services page covers the Merchant Center side.
For property portals, the useful markup is more modest: Organization for the brand, BreadcrumbList for area and building hierarchy, and accurate schema.org types that help machines understand the page even without a rich result. Holiday home operators can use Vacation rental markup where their pages meet Google's requirements. The bigger win on portals usually comes from clean HTML: a clear title, price, location and key facts in text, not inside images.
XML sitemaps and indexation monitoring for sites with tens of thousands of URLs
Split sitemaps by page type, keep only indexable canonical URLs in them, and give each an accurate lastmod. Google's sitemap documentation sets a limit of 50,000 URLs or 50MB uncompressed per sitemap, says it uses lastmod only when it is consistently accurate, and states that it ignores priority and changefreq.
Splitting by type, such as listings, areas, buildings, agents, articles, and by language, turns Search Console's sitemap report into a diagnostic tool. If listing sitemaps show a low indexed share while area sitemaps are healthy, the problem is listing quality or duplication, not the whole site.
Sitemaps must be generated from the same rules as the site. An expired listing should drop out when it starts returning 404. A filtered page blocked in robots.txt should never appear. A URL whose canonical points elsewhere does not belong. We add automated checks that fail the build or alert you if these rules break, because sitemap drift is one of the quietest ways a large site loses crawl efficiency.
How much do technical SEO services in Dubai cost?
With us, technical SEO programmes start from US$150/mo, and the final scope depends on the size and complexity of the site. Across the Dubai market, quotes for technical SEO services vary widely, and the reasons are mostly about labour and depth rather than tools.
The main cost drivers are the number of distinct templates, total URL count, whether log files are available and in what format, the rendering setup, the number of languages, and who implements fixes. An audit that stops at a report is cheaper up front, but many reports sit unread because developers cannot tell which warnings matter. A programme that includes implementation costs more per month and usually less per fix actually shipped.
When comparing proposals, ask how many of the recommendations the vendor will implement or verify, whether server logs are part of the scope, how bilingual issues are handled, and how success is measured. Indexed pages of the right type, crawl requests landing on valuable URLs, field Core Web Vitals and organic entrances to listing and category pages are all measurable. Rankings for a handful of phrases are not a fair measure of technical work, and nobody can promise them.
How a technical SEO audit and fix programme runs, month by month
Expect a diagnosis in the first two to three weeks, the highest-impact fixes shipped in the first two months, and measurable crawl and index changes after that. Google recrawls large sites gradually, so effects arrive in waves rather than all at once.
Month one is evidence: access to Search Console and logs, a full crawl, rendering comparisons per template, a hreflang check and field Core Web Vitals. The output is a ranked fix list, each item with the URLs affected, the expected effect, the code change and a test. Month two is the big structural work: status codes for expired items, parameter handling, sitemap generation, canonical and hreflang corrections. Month three onwards covers performance work, structured data, internal linking between hub and listing pages, and monitoring.
Each fix is validated after release: a spot crawl, URL Inspection on samples and, a few weeks later, the change in Search Console reports. If a release breaks something, such as a new template missing canonicals, monitoring should catch it within days rather than at the next quarterly audit.
How to choose technical SEO services in Dubai: questions to ask
Hire the team that can explain your specific problems in your own data during the sales conversation. Anyone can list generic checks; far fewer can open your Search Console, point to a pattern and explain the likely cause.
- Will you analyse our server logs, and how?
- How do you test rendering for our framework?
- How will you handle hreflang between our English and Arabic pages?
- Do you write code, or only tickets? Who validates the fix?
- What will you not touch without our approval?
- How will we measure progress in the first ninety days?
Red flags include guaranteed rankings, reports with hundreds of equally weighted warnings, recommendations to noindex large parts of the site without evidence, link packages bundled into a technical scope, and requests for your Google account password instead of a user invitation. For wider proposal-reading advice, our SEO packages page explains how to compare monthly scopes.
Technical SEO for AI search: crawler access and machine-readable pages
If you want AI search tools to cite your pages, make sure their crawlers are allowed and your content is readable without JavaScript. The same technical foundations that help Google also help AI assistants, and a few robots.txt decisions are specific to them.
Different bots do different jobs. OpenAI's documentation says OAI-SearchBot is used to surface websites in ChatGPT search features, while GPTBot collects content that may be used to train its models; blocking one does not block the other. Google says its Google-Extended token controls use of content for Gemini training and does not affect inclusion or ranking in Google Search. Many UAE sites have blanket bot rules written years ago that block useful crawlers, or allow everything without anyone deciding.
Beyond access, AI systems favour pages with clear facts in text: prices, locations, specifications, dates. Server-rendered HTML, consistent structured data and plain headings that match questions all help. We review robots.txt with you and document which bots are allowed and why, so it is a business decision rather than an accident.
Worked example: a hypothetical Dubai property portal with 120,000 URLs
Say a property portal lists homes for sale and rent across Dubai and Abu Dhabi, in English and Arabic, with roughly 120,000 URLs in its sitemaps. New listings take a week or more to appear in Google, and Arabic listing pages rarely rank. This is a hypothetical scenario to show how technical SEO services in Dubai would approach it, not a client project.
The log analysis might show that most of Googlebot's requests go to filter and sort URLs and to listings that expired months ago but still return a 200 page. The Page indexing report shows many soft 404s and many “Discovered – currently not indexed” listing URLs. A rendering check shows that listing descriptions load by JavaScript after the page shell. The hreflang audit finds the Arabic listing pages canonicalised to their English versions.
The fix plan, in order: return 410 for expired listings with links to similar live units, and drop them from sitemaps; block sort parameters and non-searched filter combinations; create static landing pages for demand-backed area and property-type combinations; server-render listing content; correct canonicals so each language version points to itself, with hreflang pairs in the sitemaps. A programme like this would start from US$150/mo, with the final scope depending on who implements the code changes.
Working with a remote technical SEO team in India from Dubai
Technical SEO works well remotely because the evidence lives in accounts and code, not on premises. India runs 1.5 hours ahead of the UAE, so a Dubai team's morning stand-up at 9:30 is 11:00 for us, and your afternoon releases can be checked the same evening.
Access comes first and stays under your control: you add us as users on Search Console and analytics, share log exports or read access to log storage, and give repository access if we are committing fixes. We never need your Google password. Fixes go through your normal review process, whether that is pull requests, a staging site or a change ticket.
The first two weeks: a scoping call over video, access set up, a full crawl and log sample, and a short written readout of the three biggest problems with the evidence for each. You can forward that readout to your developers or management before committing to the full programme. Quotes are itemised in USD; payment is by Wise, bank wire or PayPal, and invoices come from India. See our pricing page for starting prices across services.