Chrome changes how it counts in May 2027
For anyone whose team reviews a single-page app’s speed numbers every month.
In May 2027 Chrome plans to change how its public speed data counts single-page apps. If you run one, your Core Web Vitals chart may jump that month even though you changed nothing. Put a note on the chart now, while you still know why.
That public data is the Chrome User Experience Report (CrUX), gathered from real Chrome users. It has always measured full page loads. A single-page app (SPA) changes its view without loading a new page, and CrUX hasn’t counted these soft navigations.
On 2026-10-13 Chrome announced experimental soft-navigation metrics in CrUX, in a post by Annie Sullivan. The aim is to “graduate from experimental in May of 2027, subject to feedback.”
Largest Contentful Paint (LCP), a Core Web Vital, is the time until the largest element on screen appears. For Android sites where most navigations are soft, the share rated Good for LCP goes from 26% to 87% in the experimental data. She adds that “this isn’t because SPAs are faster than multi-page websites (MPAs) overall, but because follow-up clicks are faster on any site.”
Before May 2027, do 2 things.
First, mark May 2027 on the chart as a change in the counting rules. In 2023 I built a performance dashboard on our docs site, with weekly numbers per page coloured by threshold. On such a page, a sudden jump looks like an improvement unless a note explains it.
Second, decide whether you care about the first page load or the next click. After May 2027 the headline number mixes them.
A metric can move because its rule changed or because the people it counts changed. In the YouTube story, the average load time rose after the page got much faster, because users on slow connections stopped giving up. This one is a rule change.