Core Web Vitals
A small set of user-centred performance metrics covering loading, interactivity and visual stability, measured on real users rather than in a lab.
Their value is that they measure experience rather than events an engineer chose to instrument. Time to first byte is a server property; whether the main content appeared and the page responded to a tap is what a user actually noticed.
Largest Contentful Paint captures when the main content became visible, and is usually dominated by server response time, render-blocking resources and image loading. Interaction to Next Paint captures responsiveness across the whole session, and is dominated by main-thread work — which is why it punishes heavy hydration and large JavaScript bundles. Cumulative Layout Shift captures visual stability, and is nearly always caused by images and embeds without reserved dimensions, or by content injected above existing content.
The methodological point that changes how teams work: these are assessed at the 75th percentile of real users, not on a developer's machine. A site that is fast on a laptop over office broadband and slow on a mid-range Android phone over mobile data fails, correctly. Synthetic testing is useful for catching regressions in a pipeline; only field data tells you what your users experience.
The commercial hook that gets them funded: they influence search ranking, and their correlation with conversion is well documented, which turns a performance argument into a revenue one.