Data Analysis
Core Web Vitals in April 2026: The Data
For the first time in the history of the web, more than half of all origins pass Core Web Vitals. The January 2026 Chrome User Experience Report (CrUX) data shows 55.7% of origins achieving "good" scores across all three metrics -- LCP, INP, and CLS. This is a milestone two years in the making, and the data reveals where the gains came from, which platforms are leading, and where the web still struggles.
This analysis draws on CrUX data, the HTTP Archive Web Almanac 2025, and DebugBear's CWV Technology Report. All percentages represent the proportion of origins where at least 75% of page loads meet the "good" threshold for each metric.
The overall picture: Steady improvement, stubborn gaps
The web has been on a slow but consistent upward trajectory since Google introduced Core Web Vitals as a ranking signal in 2021. Mobile pass rates have improved from 36% in 2023 to 48% in 2025 -- a 12 percentage point gain over two years. Desktop has tracked slightly ahead, reaching 56% in 2025.
But the improvement is uneven across metrics. CLS showed the biggest single-year gain at +9 percentage points, driven by browser-level improvements in how layout shifts are calculated and framework adoption of automatic image dimension attributes. INP, which replaced FID in March 2024, has also improved rapidly as developers adapt their interaction patterns.
LCP remains the bottleneck. At 68.3%, it has the lowest individual pass rate of the three core metrics, and it drags down the overall number because all three must pass simultaneously. The fundamental challenge with LCP is that it depends on infrastructure (server response time), network conditions (bandwidth to download the hero image), and rendering pipeline (CSS blocking, JavaScript execution) -- a chain where every link must perform well.
Core Web Vitals Pass Rate Trend (Mobile)
Source: HTTP Archive Web Almanac 2025, CrUX January 2026. Mobile p75 data.
Per-metric breakdown: Where the web stands
LCP: The persistent bottleneck
LCP good rates rose from roughly 59% in 2023 to 62% in 2025 on mobile -- a modest +3 percentage point gain over two years. The January 2026 CrUX data shows 68.3% of origins achieving good LCP across all devices. LCP improvement is slow because the primary driver, TTFB, depends on hosting infrastructure decisions that site owners are reluctant to change.
The data reveals a stark divide: sites using CDN-first platforms (Shopify, Wix, Squarespace) achieve median TTFB under 200ms, while self-hosted WordPress sites on shared hosting average 800-1200ms TTFB. This 600-1000ms gap at the TTFB level cascades directly into LCP. Only 35% of "regular" websites use Cloudflare, compared to 70% of the top 1,000 sites -- a CDN adoption gap that directly explains the LCP disparity between top-tier and average sites.
The other major LCP bottleneck is image optimization. Only 2.1-2.2% of pages use resource preloading for their LCP image, and 16-17% of pages lazy-load their LCP image (which hurts LCP by deferring the most important image). Native loading="lazy" usage has grown from 9.5% to 10.4% on mobile, but developers are still applying it indiscriminately to hero images rather than reserving it for below-fold content.
fetchpriority="high" on the LCP image and remove loading="lazy". See our LCP optimization guide.
INP: Rapid improvement since replacing FID
INP has shown the most dramatic improvement trajectory. Since replacing FID as a Core Web Vital in March 2024, the percentage of origins with good INP rose from approximately 74% to 77% on mobile by end of 2025, and CrUX January 2026 data shows 87.1% across all devices. This rapid improvement reflects both developer awareness (INP is more demanding than FID, which motivated optimization) and framework-level improvements (React 19 transitions, Vue 3 reactivity efficiency, Angular signals).
However, the median Total Blocking Time (TBT) on mobile actually increased to 1,916 milliseconds in 2025 -- up 58% from 1,209ms in 2024. Desktop TBT also rose from 67ms to 92ms. This paradox (improving INP despite worsening TBT) suggests that while individual interactions are getting faster (likely through better event handler optimization), the overall amount of JavaScript on pages continues to grow. The long tail is getting worse even as the typical case improves.
CLS: The success story
CLS is the best-performing metric at 80.9% good origins, and it showed the biggest year-over-year improvement at +9 percentage points. This gain came from three converging factors:
- Browser-level improvements. Chrome refined its layout shift calculation in 2024-2025, reducing false positives from user-initiated scroll-based layout changes.
- Framework adoption of image dimensions. WordPress 6.5+ now automatically adds width and height attributes to images via the
wp_get_attachment_imagefunction. Next.js, Nuxt, and other frameworks have similar automatic dimension handling. - Developer awareness. CLS was the least understood metric initially, but by 2026, patterns like explicit aspect-ratio containers and font-display strategies are well-documented and widely adopted.
CMS and framework performance rankings
The technology powering a website has a significant impact on its Core Web Vitals performance. Here is how major platforms rank based on the latest CrUX and HTTP Archive data:
CWV Pass Rate by Platform (Desktop)
Source: HTTP Archive CrUX Report, November 2025 data. Desktop CWV pass rate.
The most striking finding: managed, CDN-first platforms dominate the top of the ranking. Wix, Webflow, and Shopify all serve pages from edge CDN nodes with automatic image optimization -- their users get fast TTFB and optimized assets without any manual effort. WordPress and Magento, which depend on self-managed hosting, trail significantly.
WordPress's 50% pass rate is particularly notable given that it powers 43% of the web. Because WordPress performance varies enormously by hosting tier (shared hosting averages 1.2s TTFB vs. managed hosting at 180ms), the aggregate number masks a bimodal distribution: well-hosted WordPress sites perform comparably to Shopify, while the majority on cheap shared hosting fail Core Web Vitals.
| Platform | Good LCP | Good INP | Good CLS | Good TTFB |
|---|---|---|---|---|
| Wix | 82% | 74% | 65% | -- |
| Shopify | 78% | 77% | 95% | -- |
| Squarespace | 70% | 70% | 94% | -- |
| WordPress | 50% | 46% | -- | 32% |
| Magento | 40% | 41% | 19% | -- |
Source: HTTP Archive CrUX Report, November 2025. Desktop data. Shopify's 95% CLS is the highest of any major platform.
Mobile vs. desktop: The gap persists
The mobile-desktop performance gap remains a defining feature of the web. Desktop pass rates consistently exceed mobile by 7-9 percentage points:
Desktop vs Mobile CWV Pass Rate
Source: HTTP Archive Web Almanac 2025. Overall CWV pass rate by device type.
The persistent gap is driven primarily by two factors: mobile devices have slower CPUs (affecting INP and JavaScript parse time) and mobile networks have higher latency (affecting TTFB and LCP). Even on 4G/LTE connections, the 150ms round-trip time adds measurable delay compared to broadband's 40ms RTT.
Interestingly, the gap between home pages and secondary pages is also significant. According to the Web Almanac 2025, 61% of desktop pages and 56% of mobile pages achieve good Core Web Vitals on secondary/interior pages, compared to only 47% (desktop) and 45% (mobile) on home pages. Home pages tend to have more dynamic content, larger hero images, more third-party scripts, and complex layouts -- all of which hurt Core Web Vitals.
The TTFB problem: The web's biggest structural issue
TTFB (Time to First Byte) is not a Core Web Vital -- it is a diagnostic metric. But it is the single most important upstream factor for LCP, and the data shows it is barely improving. Only 44% of mobile origins achieved good TTFB in 2025, up just +2 percentage points from 2024.
TTFB improvement is slow because it requires infrastructure changes, not code changes. Moving from shared hosting to managed hosting, adding a CDN, or switching to static generation are all significant investments. For the 43% of the web running WordPress, the WordPress core team's performance improvements cannot overcome the fundamental limitation of PHP rendering on cheap shared servers with no CDN.
The data is clear: the single highest-leverage optimization for most websites is improving TTFB through CDN adoption, page caching, or static generation. Only 35% of regular websites use Cloudflare (the most popular CDN), while 70% of top-1000 sites do. Closing this CDN adoption gap would likely move the overall CWV pass rate above 60%.
The TBT paradox: More JavaScript, faster interactions
One of the most counterintuitive findings in the 2025 data: Total Blocking Time on mobile rose 58% (from 1,209ms to 1,916ms) while INP pass rates improved. How is this possible?
The answer lies in how INP is measured versus TBT. TBT measures all long tasks (>50ms) during page load. INP measures the responsiveness of individual user interactions throughout the page lifecycle. Sites are shipping more JavaScript overall (bigger bundles, more third-party scripts), which worsens TBT. But they are also better at deferring non-critical JavaScript until after load, using code splitting to load only what is needed, and optimizing the specific event handlers that users trigger.
In other words, the web is getting heavier but smarter about when it processes that weight. The risk is that this balance is fragile -- as pages continue to accumulate JavaScript, eventually the main thread will be too congested even for well-optimized event handlers. At the 90th percentile, mobile users already face over 7.5 seconds of blocking time before full interactivity.
What this means for developers
Based on the data, here are the highest-impact actions for improving Core Web Vitals in 2026:
- Fix TTFB first. If your TTFB exceeds 800ms, no other optimization will get you to good LCP. Add a CDN, implement page caching, or switch to static generation. This is the single highest-leverage change for most sites.
- Stop lazy-loading your LCP image. 16-17% of sites still do this. Use
fetchpriority="high"on the LCP image and ensure it is in the initial server response HTML. - Preload your LCP image. Only 2% of pages preload their LCP resource. Adding
<link rel="preload" as="image">can improve LCP by 200-500ms. - Audit your JavaScript growth. TBT is rising 58% year-over-year. Use bundlewatch or Lighthouse CI budgets to prevent JavaScript creep. Every 100KB of JavaScript costs 150-300ms of parse time on mobile.
- Use your framework's built-in optimizations. Next.js Image, Nuxt Image, WordPress auto-dimensions -- these framework-level features exist to prevent the most common performance mistakes. Use them.
Methodology and data sources
This analysis draws from the following public data sources:
- Chrome User Experience Report (CrUX) -- Google's public dataset of real user experience metrics from Chrome users who have opted in to syncing their browsing history. January 2026 release.
- HTTP Archive Web Almanac 2025 -- Annual analysis of millions of websites, using July 2025 measurement data. Performance chapter covers CWV trends, rendering patterns, and resource loading.
- DebugBear CWV Technology Report -- Ongoing analysis of CWV pass rates broken down by CMS and JavaScript framework, using CrUX data.
- Google CrUX Release Notes -- Monthly CrUX releases with aggregate pass rate tracking and notable changes.
All "good" percentages represent origins where at least 75% of page views meet the good threshold for each metric (LCP <= 2.5s, INP <= 200ms, CLS <= 0.1). Mobile and desktop data are reported separately where noted; "overall" includes both device types.
Frequently asked questions
What percentage of websites pass Core Web Vitals in 2026?
As of January 2026 CrUX data, 55.7% of all web origins globally pass all three Core Web Vitals thresholds (LCP, INP, and CLS). This is up from approximately 50% in early 2024 and 53% in September 2025. Desktop pass rates are higher at 57.1% compared to 49.7% on mobile.
Which Core Web Vital is hardest to pass?
Largest Contentful Paint (LCP) remains the hardest metric to pass, with only 68.3% of origins achieving a good score. By comparison, 87.1% of origins pass INP and 80.9% pass CLS. LCP is the bottleneck for overall CWV pass rates because it depends on server response time, image optimization, and render-blocking resources.
Which CMS has the best Core Web Vitals?
Among major CMS platforms, Wix leads with an 82% desktop CWV pass rate, followed by Webflow at 79% and Shopify at 78%. WordPress trails at 50% on desktop, primarily due to TTFB issues on shared hosting. However, WordPress on managed hosting with proper caching achieves pass rates comparable to Shopify.
How much did Core Web Vitals improve from 2024 to 2026?
Mobile CWV pass rates improved from 36% in 2023 to 44% in 2024, reaching 48% in 2025. Desktop improved from 48% to 55% to 56% over the same period. The overall rate across all devices reached 55.7% by January 2026. CLS showed the biggest improvement at +9 percentage points year-over-year.
Does passing Core Web Vitals improve search rankings?
Core Web Vitals are a confirmed Google ranking signal, but they function as a tiebreaker rather than a dominant factor. Content relevance, backlinks, and topical authority remain far more important. However, page experience signals including CWV can make the difference between position 5 and position 3 for pages with similar content quality.