Core Web Vitals 2026: Los estandares de rendimiento que clasifican y convierten

In March 2024, Google replaced First Input Delay with Interaction to Next Paint as the third Core Web Vital. In 2026, the weighting of these signals in the Page Experience ranking factor has increased again. The sites that ignored the 2024 update are now seeing measurable ranking declines. The sites that implemented INP fixes early are seeing competitive gains on high-value keywords. Performance is not a developer concern that marketers can safely delegate — it is a direct revenue variable connected to both rankings and conversion rates on every page receiving paid or organic traffic.

INP: why most sites fail and how to fix it

Interaction to Next Paint measures the time between a user interaction — click, tap, keypress — and the next visual update on screen. The Good threshold is below 200ms; the Poor threshold is above 500ms. Most sites fail INP because of main thread blocking: JavaScript tasks that prevent the browser from responding to user input. The most common culprits are synchronous third-party scripts (analytics platforms, chat widgets, consent management tools, A/B testing libraries), large JavaScript bundles loaded eagerly, and long animation frames triggered by scroll or resize events. The fix hierarchy: lazy-load all non-critical third-party scripts; break up long JavaScript tasks using scheduler.postTask(); defer any script not required for initial render.

LCP: the metric Google weights most heavily

Largest Contentful Paint measures when the largest visible element on the page — usually a hero image or a headline text block — finishes rendering. The Good threshold is below 2.5 seconds. On mobile with a 4G connection, most sites with a hero image larger than 300KB fail this threshold. The most reliable LCP fixes, ranked by impact: first, preload your LCP image using a link rel=preload tag in the HTML head; second, compress your hero image to WebP or AVIF format, typically 40 to 65% smaller than JPEG at equivalent quality; third, use server-side rendering or static site generation for above-the-fold content so the browser receives ready-to-display HTML rather than an empty shell requiring JavaScript to populate.

CLS: the layout shift trap in 2026

Cumulative Layout Shift measures unexpected visual instability — elements that move after the page initially renders. The Good threshold is below 0.1. In 2026, Google's CLS measurement now includes layout shifts that occur after user interaction, not just on initial load. This catches more sites than before. The most common CLS culprits are images without explicit width and height attributes, web fonts that cause text reflow when they load, and dynamically injected content above existing content — ad units, cookie banners, notification bars. The fix: always declare image dimensions in HTML, use font-display swap with a size-adjusted fallback font, and reserve space for dynamic elements using min-height.

• Run PageSpeed Insights on your 5 highest-traffic pages — identify which Core Web Vital fails Good on mobile.
• Check Chrome UX Report in Google Search Console for real-user data, not just lab data.
• Audit your third-party scripts — remove any not generating measurable value.
• Implement resource hints (preload, prefetch, preconnect) for critical resources.

The business case: performance as a revenue line

Performance investment has a clearer ROI calculation than almost any other web investment. If your site loads in 4 seconds and a fix to 2 seconds improves conversion by 10% — a conservative estimate based on our data — on 50,000 monthly visitors at a 3% conversion rate and 150 euros average order value, the improvement generates 225,000 euros in additional annual revenue. The performance engineering work to achieve this on a typical Next.js or WordPress site costs between 5,000 and 15,000 euros. The ROI case writes itself.

Monitoring: don't fix and forget

Core Web Vitals regressions happen silently — a new marketing script gets added, a plugin update introduces a rendering issue, a new campaign landing page is built without performance review. The only defence is continuous monitoring. Set up automated Lighthouse CI in your deployment pipeline to fail builds that regress Core Web Vitals below Good thresholds. Configure Google Search Console alerts for Core Web Vitals status changes. Run a manual PageSpeed Insights audit on every new page before it goes live on a URL receiving paid traffic. Performance is not a project — it is a hygiene practice requiring the same continuous attention as security updates.