Technical SEO: The Fixes Most Dubai Websites Are Missing

There is a line in Google Search Console that quietly costs UAE businesses more traffic than any competitor does. It reads: Crawled, currently not indexed. It means Google visited the page, read it, and decided not to include it in search results at all. No penalty, no warning, no email. The page simply does not exist as far as search is concerned. It is one of the most common problems on business websites, and it tends to hit the exact service pages the business cares about most. Technical SEO is the work of making sure that never happens, and this article covers the fixes worth doing, in the order we would do them.

What Technical SEO Actually Is

Technical SEO is everything that decides whether a search engine can find your pages, read them properly, load them fast enough to be useful, and trust what they say. It is the plumbing. Content and links are what you put through the pipes.

Four things happen before you can rank for anything. A crawler has to discover the page. It has to render the page, which means running your JavaScript and seeing what a visitor would see. It has to decide the page is worth indexing. And then it has to retrieve that page as a candidate answer when somebody searches. Break any one of those four steps and nothing downstream matters. The best article ever written about aluminium fabrication will not rank if a stray tag tells Google to ignore it.

Worth being honest about the ceiling, though. Technical SEO is a qualifier, not a differentiator. Fast, crawlable and secure are baseline expectations in 2026, not competitive advantages. Fixing your technical foundation lets you compete. It does not win the race by itself. Anyone selling you technical SEO as a growth strategy on its own is selling you a foundation and calling it a house.

Fix One: Find Out What Google Has Actually Indexed

Start here, always. Open Google Search Console, go to Indexing, then Pages. You will see two lists: pages that are indexed, and pages that are not, with a reason attached to each.

The reasons that matter most:

• Crawled, currently not indexed. Google read the page and passed. Usually this means the page is thin, near-duplicate of another page, or gives no reason to exist. It is a content quality signal wearing a technical costume.
• Discovered, currently not indexed. Google knows the URL exists but has not bothered visiting yet. On small sites this points to weak internal linking. The page is an orphan that nothing on your site links to.
• Excluded by noindex tag. Someone told Google not to index it. Sometimes deliberately, often not. This is the single most expensive mistake in web development, and it happens when a site is built on a staging server with indexing switched off and goes live with the setting still on.
• Duplicate without user-selected canonical. Google found several versions of the same page and picked one for you. It does not always pick the one you want.
• Alternate page with proper canonical tag. Usually fine and intentional. Check it anyway if the URL surprises you.

Then check your sitemap. It should list only pages you want indexed, include a lastmod date that updates when you actually change content, and be submitted in Search Console. A sitemap listing 400 URLs on a 60 page site is a sign that something is generating junk pages, usually tag archives or filter parameters.

Fix Two: The Core Web Vital Nearly Everyone Fails

Core Web Vitals are three measurements of how a page feels to use. The pass marks in 2026 are largest contentful paint under 2.5 seconds, interaction to next paint under 200 milliseconds, and cumulative layout shift under 0.1.

In plain terms: how quickly the main thing on screen appears, how quickly the page responds when you tap something, and how much the layout jumps around while loading. Google replaced first input delay with interaction to next paint in March 2024, and INP is now the metric most sites fail.

That trips people up, because the usual advice is about images. Compress your images and you fix loading speed. INP is different. It measures responsiveness across the whole visit, and it is a JavaScript problem. The usual culprits are third-party scripts: live chat widgets, analytics stacks, heat mapping tools, ad pixels, cookie consent banners, and review widgets. Each one seemed harmless when it was added. Together they occupy the browser's main thread and the page stops responding to taps for a few hundred milliseconds at a time.

There is a regional edge to this too. A large share of Gulf traffic comes from mobile devices on mixed network quality, and CrUX thresholds are measured on real users, not on your office wifi with a new iPhone. Test with a mid-range Android on a normal connection and you will see what Google sees.

Fix Three: Mobile Parity, Not Just Mobile Friendly

Google indexes the mobile version of your site as the primary version. Not the desktop version, the mobile one. Whatever is missing from mobile is effectively missing from Google's view of your site.

Most sites pass the mobile friendly test and still fail on parity. The pattern is always the same: a design decision hides content on small screens. Product specifications collapse behind a tap to read more. A comparison table is removed on mobile because it did not fit. The detailed service description is desktop only, and mobile gets a two line summary. Every one of those decisions deletes content from your search presence.

The check takes two minutes. Open your most important page on a phone, and confirm that every heading, every paragraph and every image that exists on desktop is also reachable on mobile. Collapsed inside an accordion is fine. Removed from the page entirely is not.

Fix Four: Make Sure AI Crawlers Can Reach You

Crawl access used to be a Googlebot conversation. In 2026 there are six crawlers worth caring about: Googlebot, Bingbot, GPTBot, ClaudeBot, PerplexityBot and Google-Extended. The last four decide whether your business can appear inside AI written answers at all.

Here is the part that catches people. Very few businesses decide to block AI crawlers. Plenty of them do it by accident. A security plugin ships with an aggressive bot list. A CDN bot fighting mode gets switched on to stop scrapers. A developer copies a robots.txt from another project. The result is a business that is invisible to ChatGPT and Perplexity while its owner is reading articles about how to get cited by ChatGPT and Perplexity.

Check three places: your robots.txt file, your security plugin's bot blocking rules, and your CDN or firewall settings. If you want AI visibility, those user agents need to be allowed through all three.

Fix Five: One Page, One URL

Search engines treat every distinct URL as a distinct page. Most business websites accidentally publish the same page at four or five addresses.

• With and without www at the start.
• With and without a trailing slash at the end.
• http and https versions both resolving instead of redirecting.
• Filter, sort and tracking parameters creating endless variations of a category page.
• Uppercase and lowercase versions of the same path on some servers.

Every duplicate splits the signals that should be concentrated on one page. Links point at three versions instead of one. Google picks a favourite, and it is often not the one you would choose. The fix is two things done together: 301 redirects so that every variation lands on one address, and a self-referencing canonical tag on every page declaring which URL is the real one.

This matters more in this market than most, because UAE businesses rebuild their websites often. Every rebuild leaves a layer of old URLs behind, and the layers accumulate.

Fix Six: Hreflang, If You Run English and Arabic

If your site has an Arabic version, hreflang tags tell Google which version to serve to whom. Done properly, both versions rank for their own audience. Done badly, they compete with each other and split the authority you have built.

Three rules cover most cases. Use full codes rather than bare language codes, so en-AE and ar-AE rather than just en and ar. Make the tags reciprocal, meaning each version points at the other and at itself. And set an x-default for visitors who match neither. The most common failure we see is one-way tagging, where the English page points to the Arabic page but not the reverse, and Google ignores the pairing entirely.

Fix Seven: Structured Data That Matches Reality

Structured data is the layer that tells search engines and AI systems what your page means rather than just what it says. Since the March 2026 core update, Google's AI Mode reads it as a trust and entity signal rather than as a trigger for something visual, which makes accuracy matter more than volume.

For a normal business site the whole job is Organization, LocalBusiness, BreadcrumbList, and then Product or Service depending on what you sell. Anything beyond that is usually somebody padding a proposal.

What to Do This Week

If you do nothing else, do these six things in this order. All of them are free and none of them need a developer.

1. Open your robots.txt in a browser and confirm you are not blocking anything important, including the AI crawlers.
2. Open Search Console, go to Indexing then Pages, and find out how many of your money pages are not indexed and why.
3. Run your three most important pages through PageSpeed Insights and look at the field data, not the lab score. Note the INP number specifically.
4. Open your top service page on a phone and compare it line by line with the desktop version.
5. Type your homepage address five ways, with and without www, with and without the trailing slash, and over http, and confirm each one redirects to a single address.
6. Write down every finding in one list, then sort it by how much revenue the affected page could bring rather than by how severe a tool says the issue is.

None of this is glamorous work. Nobody has ever been excited about a canonical tag. But technical SEO is the only part of search where the wins are certain: fix an indexing problem and the page goes from invisible to visible, permanently, with no algorithm required to agree with you. Everything else in SEO is a negotiation. This part is just a switch that is either on or off, and it is worth an afternoon to find out which.