Core Web Vitals: LCP, INP and CLS explained for UAE sites
Core Web Vitals are three Google metrics that score how fast a page shows its main content, how quickly it reacts to a tap, and how stable its layout stays while loading.
Core Web Vitals are three metrics Google uses to measure real user experience on a web page: loading speed (LCP), responsiveness (INP) and visual stability (CLS). Google publishes a "good" threshold for each one. It judges a page by how real visitors experienced it over the last 28 days.
The three metrics and their thresholds
The thresholds below come from Google's documentation on web.dev.
| Metric | What it measures | Good | Needs improvement | Poor |
|---|---|---|---|---|
| LCP, Largest Contentful Paint | Time until the biggest visible element (hero photo, headline block) appears | up to 2.5 s | 2.5 to 4 s | over 4 s |
| INP, Interaction to Next Paint | Delay between a tap or click and the next visual response | up to 200 ms | 200 to 500 ms | over 500 ms |
| CLS, Cumulative Layout Shift | How much the layout jumps while the page loads | up to 0.1 | 0.1 to 0.25 | over 0.25 |
INP replaced First Input Delay (FID) as the responsiveness metric in March 2024. If a contractor's report still talks about FID, the report or the tool behind it is out of date.
How Google measures them
The official verdict comes from field data. These are anonymised timings from Chrome users who visited your pages, published in the Chrome UX Report (CrUX). Three details matter when you read the numbers:
- The 75th percentile. A page passes a metric when at least 75% of visits meet the "good" threshold. Your office Wi-Fi counts as one visit. A guest on mobile data in a mall car park counts as another.
- A rolling 28-day window. A fix you ship today shows up in the field numbers gradually over the next four weeks.
- Mobile and desktop are scored separately. A link from Instagram, Google Maps or WhatsApp opens on a phone, so for most local businesses the mobile result is the one to watch.
When a page has too few visits, Google groups it with similar URLs or falls back to data for the whole domain. When the whole site lacks traffic, there is no field data, and you work with lab tests such as Lighthouse.
Hypothetical example
Hypothetical example: a beach club in Jumeirah has a homepage with these elements:
- a full-screen background video
- a cookie banner
- a "Book a table" button
- a floating WhatsApp icon
Here is how that page scores on each metric:
- LCP. The largest element is the video poster frame. The server takes over a second to respond and the poster is a 3 MB JPEG. On 4G the poster appears after 4.6 s, so the page fails LCP.
- CLS. The cookie banner loads late and pushes the hero down. Then a third-party booking widget renders and shifts the menu link. A guest aiming at "Menu" taps "Book" instead. CLS lands at 0.3, which is poor.
- INP. The menu opens through a heavy script bundle. On a mid-range Android phone the first tap freezes the page for 450 ms, which needs improvement.
The fixes:
- Compress the poster to a modern format at display size.
- Reserve fixed space for the banner and the widget.
- Defer or split the script that blocks the menu.
Where Core Web Vitals matter for a UAE business
They matter wherever a visitor lands from a phone with a clear intent: book, call, check the menu, compare prices.
- Venue and restaurant sites. Hero photos and videos drive LCP. Booking widgets and chat bubbles drive CLS.
- Clinic and law firm sites. Long guide pages with embedded maps, video and consent banners tend to fail CLS.
- Car rental and real estate catalogues. Galleries, 360° viewers and filters load a lot of JavaScript and push INP up.
- Landing pages for ads. You pay for every click, and on a slow page some of those visitors leave before the offer loads.
Server speed sits underneath LCP. The Savage Media study of 537 Dubai venue websites was checked on 13.10.2026. It found:
- a median server response of 328 ms across all sites
- a median of 611 ms on WordPress sites
- 92 of 209 WordPress sites taking longer than 800 ms
A page that waits most of a second for the first byte has spent a large share of its 2.5-second LCP budget before any image starts loading. For fixes specific to Dubai hosting and themes, see the website speed guide in the Savage Media journal.
How it differs from related terms
| Term | What it is | Relation to Core Web Vitals |
|---|---|---|
| PageSpeed Insights score (0 to 100) | Lighthouse lab score from one simulated load | Lab estimate; can be high while field CWV fail, or the reverse |
| Lighthouse | Google's lab testing tool | Measures LCP and CLS in the lab; uses Total Blocking Time as a proxy for INP, since INP needs real interactions |
| TTFB, Time to First Byte | How long the server takes to start responding | Diagnostic metric that feeds into LCP; Google's "good" mark is 0.8 s; not a Core Web Vital itself |
| Page experience | Google's broader set of signals (HTTPS, no intrusive interstitials, CWV) | Core Web Vitals are one part of it |
| "Site speed" | A general phrase in sales proposals | Ask which metric, measured where, at what percentile |
What not to do
- Do not accept a lab screenshot as proof. A green 98 from one desktop run is a lab test. Ask for mobile field data from PageSpeed Insights or the Core Web Vitals report in Google Search Console.
- Do not autoplay a full-screen video on mobile without a light poster image. The poster usually becomes the LCP element, and a heavy poster can fail the metric by itself.
- Do not let widgets inject themselves into the layout. Booking, chat, review and cookie widgets that appear without reserved space are a common CLS cause on venue sites.
- Do not install a plugin for every feature. Each slider, popup and tracking plugin on WordPress adds scripts that compete for the main thread and push INP up.
- Do not judge a fix the next morning. Field data uses a 28-day window, so track the trend over a month.
- Do not chase a lab score of 100. The target is passing all three thresholds at the 75th percentile in field data. Google does not use the lab score for its verdict.
- Do not move the menu into a PDF to keep pages light. In the same study, 147 of 409 fully analysed venue sites (35.9%) offered the menu only as a PDF. Core Web Vitals do not measure the PDF, but a guest on a phone still waits for a separate heavy download.
How to check your site
- Open PageSpeed Insights, enter your homepage URL and switch to Mobile.
- Read the top block first, "Discover what your real users are experiencing". That is field data. Note LCP, INP and CLS and whether the assessment says passed or failed.
- If that block says there is not enough data, use the lab section below it. Read INP through Total Blocking Time.
- In Google Search Console, open Core Web Vitals. It sorts your URLs into good, needs improvement and poor, and names the failing metric for each group.
- Test your three money pages separately: - the homepage - the menu or catalogue - the booking or contact page
- Load each page on a real mid-range phone over mobile data. Watch for layout jumps and frozen taps.
Site structure decides which pages to fix first: the ones that carry search traffic. SEO site structure for a small business explains how to find them. Framework choice matters too. moonmars.ae is built on Astro, which outputs static HTML and ships little JavaScript by default. The website service page lists what a build includes.
Checklist
- Mobile field data checked in PageSpeed Insights, not only the lab score.
- LCP element identified (usually the hero photo or video poster) and served compressed, at display size, in a modern format.
- Server response checked. If it takes longer than 0.8 s, change the hosting or add caching.
- Fixed space reserved for cookie banners, booking widgets, chat bubbles and embedded maps.
- Width and height set on images and videos so the browser holds their space.
- Third-party scripts listed, unused ones removed, the rest deferred.
- Menu, booking and WhatsApp buttons tested by tapping on a real phone.
- Search Console Core Web Vitals report reviewed monthly, and again 28 days after each fix.
Related terms
- Google Search Console: reports field Core Web Vitals for your URLs and groups the failing pages.
- Restaurant website checklist: covers menu as a web page, online booking, WhatsApp and schema, the other basics checked in the 537-site study.
- SEO site structure: shows which pages carry search traffic and deserve speed work first.
Questions and answers
Do Core Web Vitals decide my Google ranking?
They are one of the page experience signals Google uses, and the relevance of the content carries more weight. Between two pages that answer a search equally well, the one with better Core Web Vitals has an edge.
Why does Search Console show no Core Web Vitals data for my site?
Field data comes from the Chrome UX Report, which only covers pages and sites with enough real visits from Chrome users. A new or small site often sits below that threshold, so you rely on lab tests until traffic grows.
My PageSpeed score is 95. Am I passing Core Web Vitals?
Not necessarily. The 0 to 100 score is a lab result from one simulated load, while the pass or fail verdict uses field data from real visitors over 28 days. Check the field block at the top of the PageSpeed Insights report.
Which metric should a restaurant or clinic site fix first?
Start with LCP on mobile, because heavy hero photos, background videos and slow hosting are common problems on venue sites. Then check CLS around booking and WhatsApp buttons that move while the page loads.