Skip to main content

LCP distribution

LCP distribution without noise

Georgi avatar
Written by Georgi
Updated over 3 weeks ago


What is LCP distribution

The LCP distribution lens breaks down your number of pageviews into 0.5s intervals along the ​Largest Contentful Paint (LCP) timeline. It’s an efficient way to see whether most views are fast, need improvement, or are slow in terms of loading performance.

LCP is the most popular vital from the Core Web Vitals and represents the loading time of the largest element on the page.



Healthy LCP distribution sample


Should you worry

A healthy LCP distribution looks like a declining bar chart — fewer pageviews as LCP increases. If any bar is taller than the one before it, that’s a red flag. It means more users are experiencing slower loads in that range. Focus on those spikes to identify where performance needs improvement.

Watch the P75 (75th percentile) marker too. If it is not within the green (good) zone, your LCP needs to be improved. The further to the left it is in the green area, the better.

Unhealthy LCP distribution sample

In the example below, the LCP distribution reveals a growing number of users experiencing slower load times after the 1.5s mark — a clear sign of performance issues affecting user experience.

Resolving unhealthy LCP distribution

Go-to action plan to resolve an unhealthy LCP distribution:

  1. Ask Uxi to analyze your LCP distribution and suggest improvements.

  2. Use Filters to focus on slow segments in your LCP distribution, then look across the other LCP lenses to find which ones show the slowest LCPs with the most pageviews.

  3. Simulate LCP of the suspected lens to see if fixing it will resolve the LCP distribution. If yes, this is where the resolution focus should be.

  4. Use an automated LCP optimization tool like Navigation AI to improve your LCP distribution.

  5. Once you’ve improved LCP, set an alert to be the first to know if it starts worsening again.

Try it yourself

Discover how your website performs with real user data.

Did this answer your question?