A technical SEO checklist for UAE websites

A site launched with search engines blocked is invisible however good the content is, and the check takes thirty seconds. Open yourdomain.com/robots.txt and look for a rule that blocks everything, then view the source of your home page and search it for noindex. If either turns up, stop reading and fix that first.

The rest of the list is in priority order. Each group depends on the one before, so work down and stop at the first failure. Every check says what to look at and what a bad result looks like.

First: can Google reach and index your important pages?

Google can only rank pages it can crawl and has chosen to keep, so these five checks come before everything else. Each takes minutes and needs no developer to run.

1. 01robots.txt doesn't block the site or the folders that matter. Open the file and read it. A bad result is a rule left over from a staging site that blocks everything, or one that blocks the folder holding your service pages (robots.txt).
2. 02No stray noindex. Run URL Inspection in Search Console on your five most important pages. A bad result is a noindex added by a template setting, or by the WordPress option that discourages search engines.
3. 03Every important page returns a normal response. Crawl the site or read the Pages report in Search Console. A bad result is menu links that end in 404 errors, or redirect chains of more than one hop.
4. 04The important pages are indexed (indexing). Compare the Pages report total with the number of pages you expect. 'Crawled, currently not indexed' on service pages is a quality question rather than a technical one, because Google read the page and chose not to keep it.
5. 05The sitemap lists only pages you want indexed, and is submitted in Search Console. A bad result is a sitemap full of redirected, noindexed or staging URLs (XML sitemap).

Second: does Google choose the right version of each page?

Google picks one version of every page to index, and these four checks make sure it picks the one you mean. Duplicate versions split the signals that tell Google which page matters.

1. 01One version of the site. The http address redirects to https, and the www and non-www addresses redirect to a single choice, in one hop. A bad result is two live copies of every page.
2. 02Canonical tags point where you mean. Each page's canonical should name itself unless the page is a true duplicate (canonical tag). A bad result is every page canonicalised to the home page by a plugin setting, or Arabic pages canonicalised to their English twins, which asks Google to index the English page instead of the Arabic one.
3. 03Internal links are real links. Navigation and key links should be plain links a crawler can follow, not scripts that only run on a click (internal linking). A bad result is a page reachable only through a search box or a script-driven menu.
4. 04Redirects after any change map each old page to its closest new page. A bad result is every old URL redirected to the home page after a redesign. Changing your website without losing your rankings covers the full sequence.

Third: is the site fast enough on a phone?

Google's web.dev gives three thresholds for a good experience, measured at the 75th percentile of page loads and reported separately for mobile and desktop. Mobile is the one to read first, since Google indexes the mobile version of your pages (mobile-first indexing).

Read the real-visitor data in the Core Web Vitals report before any lab score, because the thresholds describe what visitors experience. Then look first at hero images and video, sliders, font files (an Arabic font and a Latin one both have to download) and third-party tags. INP replaced First Input Delay as a Core Web Vital, so any report still scoring FID is out of date. Ask your host where the server sits relative to your visitors, and whether a CDN is in front of it.

Fourth: does the Arabic version work as its own site?

An Arabic version works as its own site when it has its own URLs, its own markup and its own layout, and these five checks test that. A pass on the English site says nothing about the Arabic one, so check each separately.

1. 01Each language has its own URL. A toggle that swaps text in place on a single URL gives Google nothing to index in Arabic. A bad result is one URL serving two languages.
2. 02hreflang pairs are reciprocal (hreflang). Each page names its twin and itself, and the twin names it back. For one country, plain en and ar are enough, and region codes such as en-AE and ar-AE are needed only when versions differ by country. A bad result is one-way tags, or invalid codes such as en-UAE, since the code for the UAE is AE.
3. 03Arabic pages declare their language and direction. The HTML language attribute should say ar and the direction should be set to rtl, the layout should mirror, and the text should be real text rather than part of an image (RTL). A bad result is Arabic baked into banners, broken punctuation, or phone numbers and English brand names flipping direction. Test on a phone first.
4. 04Arabic titles, descriptions and URLs are written in Arabic and are unique. A bad result is Arabic pages sharing the English title.
5. 05Language is a choice, not a forced redirect. If the site sends visitors to English or Arabic by location or browser setting, both versions must still be reachable by an ordinary link, because a crawler arrives without your visitors' signals. Bilingual websites in the UAE goes deeper.

What can you leave for later?

Four things attract more attention than they deserve at this stage. Each can wait until the groups above are clean.

• FAQ markup. Add structured data when it's accurate, but it won't get an ordinary business site a rich result, since Google limited FAQ rich results to well-known government and health sites in August 2023.
• llms.txt. Google's guide to generative AI features says you don't need new AI text files such as llms.txt for Search, and that content doesn't have to be broken into tiny pieces.
• Crawl budget. It matters mainly to very large sites with piles of near-duplicate URLs, such as stores with filter pages (crawl budget).
• A perfect lab score. Passing the real-visitor thresholds is the goal, not a green number in one tool.

Do the first group today. It takes an afternoon, and it decides whether the rest is worth your time. If you'd rather hand it over, our SEO service begins with a technical crawl, and what an SEO audit should cover shows what to ask for.