INP phase breakdown

The worst synthetic interaction split into input delay, processing and presentation (lab probe).

Field data PhoneDesktopAll Scope All sites Q2 2026 edition · All devices field outcomes
Metric LCP INP CLS
1

At a glance the headline numbers for INP phase breakdown

The worst synthetic interaction split into input delay, processing and presentation (lab probe).

49ms
Median INP stack
the measures at p50
96%
Biggest: Presentation
47ms of the stack
151ms
Headroom at p50
vs the 200ms target

On the typical site, the worst interaction waits 0ms before any code runs.

The State of Web Vitals · Q2 2026 · 189,915 sites · all devices field datacorewebvitals.io/state-of-cwv
2

Where the milliseconds go the phases at the median, against the budget

Input delay0ms0% of the total Processing2ms4% of the total Presentation47ms96% of the total 200ms target
Input delay 0ms · 0% Processing 2ms · 4% Presentation 47ms · 96%

At the median the whole stack uses 49ms, inside the 200ms INP budget. The biggest piece is Presentation at 47ms, 96% of the total. computed

The State of Web Vitals · Q2 2026 · 189,915 sites · all devices field datacorewebvitals.io/state-of-cwv
3

Why this matters for the Core Web Vitals, and where to start fixing it

An interaction spends time in three places. Input delay is the wait before your handler even starts, because the main thread was busy with something else. Processing is the handler running. Presentation is the wait after the handler, while the browser recalculates styles, lays out and paints the result.

Each phase has its own causes. Input delay points at long tasks from scripts that have nothing to do with the click. Processing points at a handler doing too much synchronously. Presentation points at the rendering cost of the update, which scales with DOM size. Fix the phase the data names. The other two will not move.

Related signals CSS hints → Scheduling API use → Interaction invoker types → Yielding strategy → Chrome field data from 189,915 sites, representing millions of real page loads · How we measured