Website Speed Optimization for MENA: Why Most Jordanian Sites Fail Core Web Vitals and How to Fix It

Website speed in MENA is a business problem as much as a technical one. Google's ranking algorithm uses Core Web Vitals as a ranking signal — sites that fail the 'Good' threshold rank lower than equivalent sites that pass it. And MENA mobile networks, while improving, create a performance testing environment that is harsher than the fast connections developers typically use during testing. The result: a website that feels fast during development on a UK or US fibre connection performs badly for the Jordanian customer on a 4G connection in Amman — and ranks lower in Google as a result.

The 6 most common performance failures in MENA websites

• Unoptimised images — Large JPEG or PNG images served at full resolution regardless of screen size. Fix: convert to WebP or AVIF, implement responsive images with `srcset`, use lazy loading for below-fold images.
• Unoptimised Arabic fonts — Full Arabic font files (80–150 KB) loaded synchronously, blocking rendering. Fix: subset the font to only the Unicode ranges used on the page, use `font-display: swap`, preload critical fonts.
• Render-blocking JavaScript — Large JavaScript bundles loaded in the `` that block the browser from rendering visible content. Fix: defer non-critical JavaScript, split code into smaller bundles, use dynamic imports.
• No CDN for Middle East delivery — Hosting on a server in Europe or the US without a CDN means every asset request travels a long geographic distance, adding latency. Fix: implement Cloudflare (free tier) or AWS CloudFront with MENA edge locations.
• Cumulative Layout Shift from late-loading fonts or images — Content shifts as fonts or images load, failing CLS. Fix: explicitly specify dimensions for all images, use `font-display: optional` for non-critical fonts, reserve space for dynamically loaded content.
• No caching headers — Static assets (CSS, JS, images) served without cache headers force the browser to re-download them on every visit. Fix: set `Cache-Control` headers with long max-age values for versioned static assets.

How to measure correctly for MENA

Running a Lighthouse audit on a fast desktop connection does not reflect MENA user experience. The correct measurement methodology: use Google PageSpeed Insights (which measures against real-world Chrome User Experience Report data); test using Lighthouse with the 'Mobile' device setting and 'Slow 4G' throttling; and use WebPageTest with a Middle East test location (Dubai is available) to simulate actual geographic and network conditions. The 'Good' thresholds — LCP under 2.5 seconds, INP under 200ms, CLS under 0.1 — should be achieved under mobile/slow 4G conditions, not just on desktop fast connection.

The performance audit process

A structured performance audit for a MENA website: run PageSpeed Insights and capture baseline scores by device; identify the specific Core Web Vitals metrics failing and which elements are causing the failure (PageSpeed Insights shows exactly which element is the LCP element, which resources are render-blocking, and which elements are shifting); prioritise fixes by estimated impact — image and font optimisation typically produces the largest LCP improvement; implement fixes in a staging environment; remeasure to confirm improvement before deploying to production. As the Web Development Agency in Amman Jordan, we include a performance audit and optimisation pass as a standard phase of every website build — because a slow website in MENA is both a ranking problem and a conversion problem, and fixing performance after launch is significantly more expensive than building for performance from the start.