--- title: "Core Web Vitals in Latin America: 2026 data by country" description: "How many sites pass Core Web Vitals on mobile in Mexico, Colombia, Brazil, Argentina, Chile and Peru vs the US, why mobile fails, and what fixes it." url: "https://dardo.studio/en/blog/core-web-vitals-latin-america/" language: "en" translations: {"es":"https://dardo.studio/es/blog/core-web-vitals-latinoamerica/","pt":"https://dardo.studio/pt/blog/core-web-vitals-brasil/"} updated: "2026-10-15T13:51:03.026Z" --- [Blog](https://dardo.studio/en/blog/) # Core Web Vitals in Latin America: how many sites pass in each country, and why mobile fails (2026 data) In September 2026, a third of sites passed Core Web Vitals for mobile visitors in Colombia and Brazil, against 63% in the US. Data by country and what to do. By [Nicolás Cerón](https://dardo.studio/en/nicolas-ceron/) ·October 15, 2026 ![A steep Andean hillside city at dusk with thousands of amber windows, crossed by an aerial cable car line; one crimson cabin glides ahead of the dark slate ones.](https://dardo.studio/_astro/01M4ZX8ZX3XWAGATB23B8SDSBS_Z2gknak.webp) ## Key takeaways - In September 2026, 32.4% of sites passed Core Web Vitals for mobile visitors in Colombia, 33.7% in Brazil and 40.5% in Mexico, against 63.3% in the United States. - Latin American sites aren't heavier than US sites; the median US page is the heaviest. The gap is in visitors' phones and connections: responsiveness (INP) and server response fail on mobile. - On desktop, about 95% of sites have a good INP in every country; on mobile it falls to 55–66% in most of Latin America. - The most popular sites aren't faster: in Brazil, 27.3% of the top 10,000 sites passed on mobile, against 33.6% of the top million. - Every country improved over the year, by 2 to 5 points; a site built for Latin American phones needs a smaller JavaScript budget and a server close to its visitors. ## The short answer In September 2026, about one site in three passed Google's Core Web Vitals for visitors on mobile in Colombia (32.4%), Brazil (33.7%) and Peru (29.1%), and four in ten in Mexico (40.5%). In the United States, it was 63.3%. That's real-user data from the Chrome UX Report, published by the HTTP Archive. The gap isn't because Latin American sites are built worse. The median page that visitors in Colombia or Mexico load is lighter than the median page in the US, and the two score about the same in a lab test. What differs is the visitor's phone and connection: on mobile, sites in Latin America fail on responsiveness (INP) and on how fast the server answers (TTFB), while on desktop the gap almost disappears. So a site that's "fast enough" for US visitors can still fail for most visitors in Bogotá, São Paulo or Lima. Below is the data by country and by platform, and what to change if your customers are in Latin America. You can explore it by country, platform and metric in our free [website platform speed report](https://dardo.studio/en/tools/website-platform-speed-report/). ## How many sites pass Core Web Vitals in each country Each row counts the sites (origins) that real Chrome users in that country visited often enough to be measured, and the share that had a good experience. A site passes Core Web Vitals when, at the 75th percentile over 28 days, its Largest Contentful Paint (LCP) is 2.5 seconds or less, its Interaction to Next Paint (INP) 200 milliseconds or less and its Cumulative Layout Shift (CLS) 0.1 or less. | Country | Pass on mobile | Pass on desktop | Mobile, a year earlier | Good LCP (mobile) | Good INP (mobile) | Good TTFB (mobile) | Sites measured | | ------------- | -------------- | --------------- | ---------------------- | ----------------- | ----------------- | ------------------ | -------------- | | United States | 63.3% | 65.4% | 60.6% | 74.8% | 87.3% | 54.8% | 1,288,504 | | Spain | 53.1% | 64.2% | 51.1% | 66.4% | 79.5% | 46.3% | 335,434 | | Mexico | 40.5% | 55.4% | 36.3% | 54.4% | 66.4% | 39.9% | 224,290 | | Chile | 38.5% | 50.2% | 35.0% | 48.6% | 75.6% | 27.6% | 98,860 | | Argentina | 37.9% | 52.8% | 33.2% | 50.1% | 65.1% | 31.7% | 179,755 | | Brazil | 33.7% | 55.4% | 31.7% | 48.3% | 55.3% | 38.8% | 598,869 | | Colombia | 32.4% | 50.4% | 27.0% | 45.5% | 62.9% | 32.2% | 132,188 | | Peru | 29.1% | 46.7% | 25.2% | 41.8% | 60.7% | 24.4% | 82,502 | | All countries | 52.8% | 60.8% | 50.5% | 65.6% | 79.4% | – | 8,569,446 | Source: HTTP Archive technology report (Chrome UX Report), September 2026 release, all technologies. "A year earlier" is the September 2025 release. TTFB (time to first byte) isn't one of the three Core Web Vitals; it's shown because it explains part of the LCP gap. Three things stand out: - **Mobile is where Latin America falls behind.** On desktop, Mexico (55.4%) and Brazil (55.4%) are within ten points of the US. On mobile, they're 23 and 30 points behind. - **Responsiveness is the widest gap.** On desktop, about 95% of sites have a good INP in every country in the table. On mobile, it's 87.3% in the US and between 55.3% (Brazil) and 66.4% (Mexico) in most of Latin America; Chile (75.6%) is the exception. - **Every country improved.** Colombia gained 5.4 points in a year, Argentina 4.7 and Mexico 4.2. ## Latin American sites aren't heavier; the phones and networks are slower The HTTP Archive also tests the home pages of these sites in its own lab, from the same machines for every country. The median page that Latin American visitors load is lighter than the median page US visitors load, and it scores the same or better in Lighthouse: | Country | Median page weight (mobile) | Median JavaScript (mobile) | Median Lighthouse performance (mobile) | | ------------- | --------------------------- | -------------------------- | -------------------------------------- | | United States | 2.96 MB | 1.10 MB | 48 | | Spain | 2.60 MB | 0.78 MB | 46 | | Mexico | 2.62 MB | 0.85 MB | 53 | | Chile | 2.69 MB | 0.84 MB | 50 | | Argentina | 2.24 MB | 0.73 MB | 53 | | Brazil | 2.32 MB | 0.90 MB | 50 | | Colombia | 2.53 MB | 0.83 MB | 53 | | Peru | 2.52 MB | 0.78 MB | 53 | Same lab, similar pages, very different field results. That points to the conditions of the visit: the phones people use (a mid-range Android phone takes longer than a recent flagship to run the same JavaScript, and that delay is what INP measures) and the network path to the server (TTFB is good for only 24–40% of sites on mobile in Latin America, against 55% in the US, because many sites are hosted in the US or Europe and mobile networks add latency). Two caveats. The country of a visit is inferred from the visitor's IP address, so a "Colombia" row describes what visitors in Colombia experienced, wherever the site is hosted. And this is correlation: the data shows where the gap is (mobile INP, mobile TTFB and LCP), not every cause. ## Popular sites aren't faster You might expect the most visited sites, with the biggest budgets, to do better. They don't: | Country | Top 10,000 sites | Top 100,000 | Top 1,000,000 | | ------------- | ---------------- | ----------- | ------------- | | United States | 55.1% | 59.2% | 63.9% | | Mexico | 38.5% | 40.7% | 40.5% | | Colombia | 34.1% | 33.3% | 32.4% | | Brazil | 27.3% | 27.0% | 33.6% | Share of sites passing Core Web Vitals on mobile, by popularity rank, September 2026. Large sites tend to carry more scripts: tag managers, personalization, chat, A/B testing and ads, each one a cost to INP on a mid-range phone. ## Platforms in Latin America: the same order, lower numbers The platform a site is built on changes the odds, but the country changes them more. Share of sites passing Core Web Vitals on mobile, September 2026: | Platform | US | Mexico | Colombia | Brazil | Argentina | Chile | Peru | Spain | | --------- | ----- | ------ | -------- | ------ | --------- | ----- | ----- | ----- | | Wix | 88.7% | 67.3% | 65.7% | 67.3% | 66.3% | 78.0% | 58.8% | 81.1% | | Shopify | 85.5% | 64.6% | 52.8% | 51.5% | 41.3% | 69.1% | 49.3% | 67.0% | | Astro | 84.2% | 54.7% | 46.0% | 44.5% | 45.4% | 57.0% | 46.2% | 76.0% | | Framer | 81.7% | 54.7% | 36.2% | 56.5% | 47.9% | 62.2% | 45.5% | 67.6% | | Webflow | 80.2% | 50.1% | 43.6% | 45.4% | 41.5% | 52.5% | 44.4% | 59.7% | | WordPress | 62.0% | 31.6% | 24.7% | 32.8% | 27.6% | 28.8% | 22.8% | 45.7% | | Next.js | 51.6% | 27.3% | 21.7% | 19.9% | 20.1% | 27.5% | 17.2% | 39.3% | A WordPress site in the US (62.0%) is more likely to pass than a Webflow, Framer or Astro site in Colombia or Brazil. Local e-commerce platforms do well where they're strong: in Brazil, 78.8% of Nuvemshop sites and 65.9% of Tray sites passed; in Colombia, 87.1% of Tiendanube sites did. VTEX sites passed rarely (3.5% in Colombia, 5.7% in Brazil). The [platform report](https://dardo.studio/en/tools/website-platform-speed-report/) has all 31 platforms, every country and the twelve-month trend. ## What to do if your visitors are in Latin America - **Measure where your visitors are.** PageSpeed Insights and Search Console show your field data for all visitors together; Chrome's country-level data is only in the [CrUX BigQuery dataset](https://developer.chrome.com/docs/crux/bigquery). If most of your traffic is in the US, a passing score can hide failing pages for your Latin American customers. Your own analytics, split by country, tells you how much of the audience that is. - **Budget JavaScript for a mid-range Android phone.** INP is the widest gap. Remove or delay third-party scripts, split long tasks, and test with CPU throttling in Chrome DevTools rather than on a recent laptop. Google's guide to [optimizing INP](https://web.dev/articles/optimize-inp) explains how to find the script behind a slow interaction. - **Serve from close to the visitor.** A CDN with points of presence in the region cuts the round trips that hurt TTFB and LCP: Cloudflare, for example, lists data centers in [Bogotá, Medellín, Mexico City, São Paulo, Lima, Santiago and Buenos Aires](https://www.cloudflare.com/network/). Cache HTML at the edge where you can, not only images. - **Make the largest element cheap.** Size and compress the hero image, preload it, and avoid sliders and video heroes on mobile. Google's [LCP guide](https://web.dev/articles/optimize-lcp) breaks the time into its parts so you can see which one is long. - **Fix the site before you move it.** Platforms differ, but a slow site is usually slow because of what was added to it. Our [Core Web Vitals playbook](https://dardo.studio/en/blog/pass-core-web-vitals/) starts from field data and the failing template. Moving makes sense when the platform blocks the fixes; then plan the [migration](https://dardo.studio/en/services/website-migration/) so rankings carry over. Dardo is a studio in Bogotá; we test on the phones and connections our Latin American clients' customers use. If your site has to work for visitors in the US and Latin America, our [speed optimization service](https://dardo.studio/en/services/website-speed-optimization/) starts with your field data by country. ## How this data was collected - **Field data** comes from the [Chrome UX Report](https://developer.chrome.com/docs/crux/methodology) (CrUX): anonymous measurements from Chrome users who opted in, aggregated per origin over 28 days. Only sites with enough visits are included, and the country-level data applies that threshold per country. - **Countries** are [inferred from visitors' IP addresses](https://developer.chrome.com/docs/crux/methodology/dimensions). - **Platforms, page weight and Lighthouse scores** come from the HTTP Archive's monthly crawl, which detects technologies on each site's home page and one inner page and tests home pages in its lab. - **The numbers** were read from the HTTP Archive's public technology report API on 15 October 2026 (the September 2026 release). Our [platform report](https://dardo.studio/en/tools/website-platform-speed-report/) reads the same API on every build, so it shows the latest month. ## Common questions **What percentage of websites pass Core Web Vitals in Latin America?** In September 2026, on mobile: 40.5% in Mexico, 38.5% in Chile, 37.9% in Argentina, 33.7% in Brazil, 32.4% in Colombia and 29.1% in Peru, against 63.3% in the United States and 52.8% across all countries. On desktop, between 46.7% (Peru) and 55.4% (Mexico and Brazil). **Why do websites fail Core Web Vitals more often in Latin America?** Mostly because of the visit, not the site: on mobile, fewer sites have a good responsiveness (INP) and server response (TTFB) in Latin America, while on desktop the numbers are close to the US. Mid-range phones run JavaScript more slowly, and many sites are served from far away. The median page Latin American visitors load is actually lighter than the US median. **Which Core Web Vital fails most in Latin America?** LCP fails most often (between 41.8% and 54.4% of sites have a good LCP on mobile in the region), and INP shows the widest gap with the US: 55–66% of sites have a good INP on mobile in most Latin American countries, against 87% in the US. **How can I see my site's Core Web Vitals for visitors in one country?** PageSpeed Insights and Search Console don't split field data by country; the CrUX BigQuery dataset does, for sites with enough visits in that country. For a smaller site, your own analytics with web-vitals measurement split by country is the practical option. **Which website platform is fastest in Latin America?** Among platforms with enough sites, hosted builders and local e-commerce platforms lead: in Colombia Tiendanube and Wix, in Brazil Nuvemshop, Wix and Tray. Next.js and WordPress sites pass least often in every country. Compare platforms by country in our free platform report. **Is it getting better?** Yes. From September 2025 to September 2026, the share of sites passing on mobile rose by 5.4 points in Colombia, 4.7 in Argentina, 4.2 in Mexico, 3.9 in Peru, 3.5 in Chile and 2.0 in Brazil. ## Sources & further reading Sources behind this page, with further detail from the original publishers. - [HTTP Archive: Technology Report](https://httparchive.org/reports/techreport/landing) · httparchive.org - [HTTP Archive: Technology Report APIs (GitHub)](https://github.com/HTTPArchive/tech-report-apis) · github.com - [Chrome for Developers: CrUX methodology](https://developer.chrome.com/docs/crux/methodology) · developer.chrome.com - [Chrome for Developers: CrUX dimensions (country)](https://developer.chrome.com/docs/crux/methodology/dimensions) · developer.chrome.com - [Chrome for Developers: CrUX on BigQuery](https://developer.chrome.com/docs/crux/bigquery) · developer.chrome.com - [web.dev: Web Vitals](https://web.dev/articles/vitals) · web.dev - [web.dev: Optimize Interaction to Next Paint](https://web.dev/articles/optimize-inp) · web.dev - [web.dev: Optimize Largest Contentful Paint](https://web.dev/articles/optimize-lcp) · web.dev - [web.dev: Optimize Time to First Byte](https://web.dev/articles/optimize-ttfb) · web.dev - [Chrome for Developers: Throttling in DevTools](https://developer.chrome.com/docs/devtools/settings/throttling) · developer.chrome.com - [Cloudflare: Global network](https://www.cloudflare.com/network/) · cloudflare.com - [Google Search Central: Core Web Vitals and Google Search results](https://developers.google.com/search/docs/appearance/core-web-vitals) · developers.google.com [Nicolás Cerón](https://dardo.studio/en/nicolas-ceron/) Nicolás Cerón is the founder of Dardo, a brand, web design and development studio in Bogotá, Colombia. ## Related articles - [**Webflow vs WordPress in 2026: real costs, editing, SEO and speed**What a bilingual 20-page site with a 300-post blog costs over three years on Webflow and WordPress, and how editing, SEO, speed and security compare.](https://dardo.studio/en/blog/webflow-vs-wordpress/) - [**WordPress to Astro: how to migrate a WordPress site, headless or all the way, without losing rankings**Two ways to move from WordPress to Astro, what to export, how to keep Yoast titles, permalinks and uploads, and the WordPress URLs most migrations forget.](https://dardo.studio/en/blog/wordpress-to-astro/) - [**Webflow to Astro: how to migrate a Webflow site without losing content, design or rankings**Webflow's code export leaves out the CMS, forms, search and localized pages. What to take out of Webflow, what to rebuild in Astro and how to keep rankings.](https://dardo.studio/en/blog/webflow-to-astro/) - [**WordPress or a custom website? How to choose, and when each one pays off**When WordPress is the right choice, when a custom site pays off, and the middle path of headless WordPress, compared on cost, speed, updates and ownership.](https://dardo.studio/en/blog/wordpress-vs-custom-website/) ## Keep exploring. - [Service · **Website speed and Core Web Vitals optimization**](https://dardo.studio/en/services/website-speed-optimization/) - [Your project · **Start a conversation**](https://dardo.studio/en/contact/)