Always-on monitoring of Largest Contentful Paint, mobile + desktop, real users + lab. Sentry catches LCP regressions, missing preloads, lazy-loaded hero images, and render-blocking before they cost you a Page Experience signal. Cortex handles the fix.
Continuous audits of mobile + desktop LCP against Google's 2.5s threshold and the documented diagnostics that explain why it passes or fails. Each rule maps to a primary web.dev or Google Search Central doc. Sentry catches what fails. Cortex fixes it.
CrUX 28-day p75 LCP on phone form factor under Google's 2.5s 'Good' threshold. Mobile field LCP is the metric Google uses for the mobile Page Experience signal.
CrUX 28-day p75 LCP on desktop under Google's 2.5s 'Good' threshold. Desktop field LCP is the metric Google uses for the desktop Page Experience signal.
Lighthouse mobile-emulated LCP from PSI under 2.5s. Lab LCP is deterministic - it doesn't depend on real user traffic - and surfaces regressions before field data updates.
Server response time (Time to First Byte) under Google's documented 800ms 'Good' threshold. TTFB is one of LCP's four sub-parts per web.dev - the front-end cannot recover an LCP budget the server has already burned through.
The LCP element appears in the server-rendered HTML, not injected by client-side JavaScript. Client-rendered LCP elements cliff on slow networks and CPUs because the browser must finish hydration before it can paint.
Zero synchronous `<script>` or unscoped `<link rel='stylesheet'>` in the head blocking the LCP critical path. Each render-blocking resource extends 'element render delay' - one of the four LCP sub-parts per web.dev.
If the LCP element is an image, a `<link rel='preload' as='image'>` tag in the head fetches it before render-blocking CSS finishes. Typically saves 100-400ms on LCP.
If the LCP element is an image, the `fetchpriority='high'` attribute marks it as priority during browser resource prioritization. Required for the modern preload-less LCP optimization path.
If the LCP element is an image, it does NOT carry `loading='lazy'`. Lazy-loading the LCP image defers its fetch past first paint and guarantees the metric fails.
If the LCP element is an image, it's served as WebP or AVIF (or a `<picture>` with one of those as the primary source). Modern formats reduce LCP payload 25-50% vs JPEG/PNG at equivalent quality.
If the LCP element is an image, both `width` and `height` attributes are set on the `<img>` tag. Explicit dimensions let the browser reserve space and avoid a reflow that delays paint and inflates CLS.
Paste your homepage URL. Sentry runs PSI for mobile + desktop, pulls 28-day CrUX field LCP for both form factors, identifies the LCP element and its preload + fetchpriority + lazy posture, and ships a per-rule report. No signup, instant results, always free.
Sentry fetches your site, runs every LCP rule, and renders the full result page before your next sip of coffee.
Each failed rule ships with a prescription paragraph. Hand it to engineering and the gap is closed before lunch.
Add your site to the daily Sentry sweep with one click. New regressions get caught the next morning.
11 rules in the LCP Sentry. Daily 3:30 AM ET sweep.
One brain. Thirty-six pairs of eyes. Sentries monitor every visibility signal that decides whether search engines, AI engines, and ad platforms show you. Cortex reads what they see, weighs it against a unified corpus of platform documentation, and acts. Every move follows a defined decision protocol: action stated, reason given, impact named.