What is the relationship between INP and the former FID metric?
In short and in plain language
Google measures how fast a page reacts to clicks. There is a new measure for this. It took over from an older one.
The old measure only looked at the first click. That test was easy to pass.
The new measure checks all clicks and inputs during a visit. It reports the worst value.
So a menu, a filter or a form must all run smoothly. We find long tasks in the browser and split them up.
INP (Interaction to Next Paint) replaced the older FID (First Input Delay) metric as a Core Web Vital in March 2024. FID measured only the delay of the very first interaction, and only up to the start of the response, not the visible result. That was an easy test to pass and said little about the real interaction experience. INP, by contrast, measures all interactions across the entire page visit and reports the worst value (minus statistical outliers).
For optimization this shifts the focus considerably: it is not enough for the page to react on the first click. Opening a menu, filtering a product list or submitting a form must all stay fluid too. To achieve that, we analyze long JavaScript tasks that block the main thread and break them up. More on this topic on the frontend optimization page.