Largest Contentful Paint, or LCP, measures when the largest eligible visible content element is rendered in the viewport. It often corresponds to a hero image or prominent text block. Improving LCP begins by identifying the actual measured element and the delay that prevents it appearing promptly.
Separate the stages of loading
LCP is one of the Core Web Vitals. A good field result is 2.5 seconds or less at the 75th percentile. Server response, discovery of the resource, download time and rendering delays can all contribute. Reducing image bytes may help when transfer is slow, but it will not resolve a long server delay or a hidden element waiting on JavaScript. The metric is useful evidence about loading; it does not measure whether the headline answers a searcher’s question.
Identify what arrives late
Use performance tooling to locate the LCP element on representative mobile and desktop pages. Inspect the request waterfall and confirm when its resource is discovered. For a main image, review dimensions, encoding, responsive variants and priority; avoid unnecessary lazy loading above the fold. For a text element, investigate fonts and blocking styles. Compare repeat visits with first visits so caching does not conceal a cold-load problem. Make one targeted change at a time and record what portion of the delay it addresses.
Check whether the candidate element differs between page variants. A promotional banner, cookie state or narrower viewport may change which content becomes largest, so one successful trace should not represent every visit.
Fix the bottleneck on a service page
Suppose a Perth builder’s main project photograph is inserted only after a large slideshow script initialises. Re-encoding that photograph alone may save bytes while leaving the main delay intact. Exposing the first image in the initial HTML and reducing the slideshow’s startup work could be a better test. Preserve an appropriate image size and visual quality. Afterwards, review the actual LCP element again because a layout change can change what is measured, and monitor field trends without promising an immediate search improvement.
An LCP investigation can be scoped through technical SEO services.
Common questions
Should every image be preloaded?
No. Excessive priority hints can compete for bandwidth. Focus on the verified important resource and confirm the effect in the loading waterfall.
Why is LCP different on my phone?
Screen size can change the measured element, while device speed, connection, cache and location affect timing. Compare representative conditions and field data.
Terms in this guide
- Largest Contentful Paint — A loading measure that reports when the largest visible content element is rendered. Large hero images, slow delivery and render-blocking resources can delay it.
- Core Web Vitals — A set of user experience measures covering loading, responsiveness and visual stability. The current metrics are Largest Contentful Paint, Interaction to Next Paint and Cumulative Layout Shift.
- Lazy loading — Deferring an asset until it is approaching the area the visitor is viewing. Main above-the-fold imagery should be prioritised rather than delayed unnecessarily.
- JavaScript — A programming language used for interactive website behaviour and application logic. Essential content should remain accessible if an optional visual feature fails.
Related SEO guides
Apply this to your website
Explore Technical SEO, SEO Audits for help planning and implementing the relevant changes, or discuss your website with Refined IT.