What is Core Web Vitals?
A set of Google page-experience metrics measuring how quickly a page loads, how fast it responds to input, and how much the layout shifts while loading.
LCP under 2.5s · INP under 200ms · CLS under 0.1
A worked example
A WooCommerce product page whose main image finishes rendering at 4.1 seconds fails Largest Contentful Paint. On a mid-range Android phone over a patchy mobile connection (which describes a large share of Indian e-commerce traffic) that page is losing sales regardless of where it ranks.
Why it matters
The direct ranking effect of Core Web Vitals is modest and Google has said so. The conversion effect is not modest, and on Indian mobile connections it is frequently the larger commercial issue. Every additional second before a page becomes usable costs you a measurable share of visitors, and on paid traffic you have already paid for every one of them.
The three metrics measure different failures. Largest Contentful Paint measures how long before the main content appears, usually a hosting, image or render-blocking resource problem. Interaction to Next Paint measures responsiveness when someone taps (usually too much JavaScript. Cumulative Layout Shift measures how much the page jumps while loading) usually images without dimensions or fonts loading late, and it is the one that makes people tap the wrong thing and leave.
For WooCommerce and WordPress sites specifically, the usual culprits are predictable: too many plugins, an unoptimised theme, uncompressed images, and shared hosting that cannot cope.
The nuance most people miss
Lab scores and field data are not the same thing, and only one of them counts. PageSpeed Insights gives you both: a lab score from a simulated test, and field data from the Chrome User Experience Report reflecting what real visitors actually experienced. It is entirely possible to score well in the lab while failing in the field, particularly if your real users are on slower devices and networks than the simulation assumes. Optimise against field data. And do not ignore Interaction to Next Paint, which replaced First Input Delay and is considerably harder to pass on plugin-heavy sites.
Common mistakes
- Optimising the lab score while field data stays poor
- Ignoring INP because the older FID metric used to pass
- Installing a caching plugin and treating the problem as solved
- Testing only on a fast desktop connection, which hides the problem entirely