Core Web Vitals

Core Web Vitals are three metrics for the quality of a webpage from the visitor’s perspective. They answer three simple questions. How long does it take for the most important content to become visible? How quickly does the page respond when you click or tap something? And does the content shift while loading, causing you to accidentally click on the wrong thing? Google established these three metrics in 2020 and measures them using real visitors. The name roughly means “central vital signs of the web,” modeled on pulse and blood pressure at the doctor’s.

Why Google turned loading times into a ranking question

Anyone who clicks on a page and stares at a blank white screen for three seconds often just goes back. For online shops, that’s directly costly: every additional second of waiting measurably costs orders. Large retailers estimate a loss of a few percent in revenue per second. That’s why site operators already have their own good reasons to pay attention to these metrics.

On top of that comes pressure from Google. Since 2021, Core Web Vitals have factored into search results as a ranking signal, meaning the order in which results are displayed. Their weight, however, is small. A slow page with the best content still beats a fast page with poor content. Only when two pages are similarly good in terms of content can speed tip the balance.

A common misconception is that good scores automatically lead to better rankings. That’s not quite true. The metrics function more like a hygiene factor: very poor results hurt, but very good ones bring no bonus.

The three metrics in detail

The first metric is called Largest Contentful Paint, or LCP for short. It measures when the largest visible element has finished loading, usually an image or a heading. Under 2.5 seconds is considered good, over 4 seconds is considered poor. It can be improved mainly through smaller image files and faster servers.

The second metric is called Interaction to Next Paint, or INP for short. It measures the delay between a click and the page’s first visible response. Under 200 milliseconds is good. Long delays usually occur because the browser is busy running code and can’t process the click immediately. INP replaced the older metric First Input Delay in March 2024, because that one only looked at the very first interaction.

The third metric is called Cumulative Layout Shift, or CLS for short. It measures how much the page content still jumps around while loading. A typical case: you want to tap on a link, an ad loads in above it, everything shifts downward, and you end up hitting the ad instead. CLS is not a time value but a ratio. Under 0.1 is considered good. This jumping can be avoided if developers reserve fixed placeholder space for images and ads from the start.

Where these metrics show up in everyday life

Anyone can check these metrics themselves. Google provides the free tool PageSpeed Insights for this: enter the address, wait briefly, get a result in green, yellow, or red. The Chrome browser also includes corresponding measurement features. Anyone managing their own project, such as a school website, will quickly see the biggest bottlenecks there.

It’s important to understand the difference between two types of data. Lab measurements are taken under artificial conditions on a test machine. Field data comes from real visitors using their actual devices and connections. Only field data counts for search results. That’s why a page can shine in lab testing and still fail in practice, for instance because many users have older phones with weak mobile connections.

In the news, the term usually comes up in the context of search engine optimization and online retail. Agencies advertise with better scores, and shop systems like Shopify or WordPress providers publicly compare themselves on this basis. The debate over online advertising is also tied to this: many ad scripts noticeably worsen several of the three metrics at once.