Single-page applications have been judged in Chrome field data almost entirely on their first load. The Chrome UX Report (CrUX) release published October 13 starts to change that, adding experimental Core Web Vitals that count in-app navigations one by one.

Search Engine Journal reported on the announcement, which came from a Chrome post written by Annie Sullivan. The experimental data can be pulled through BigQuery, the CrUX API, or the History API, and CrUX Vis gained a toggle to display it. Stable scores, the ones Google currently reports, are unchanged.

The release adds three fields: an experimental LCP field with a softnavs suffix, plus matching INP and CLS fields. Each measures the same thing as its stable counterpart. The difference is scope, since the experimental versions include both full page loads and detected soft navigations.

Until now, an SPA got credit for one page view. Everything a visitor did after landing, including later interactions and layout shifts, was charged to the landing URL. Under the new approach, any user action that changes content and URL is recorded separately and reported under the destination URL, provided it meets the usual eligibility rules.

Chrome relies on the Soft Navigations API for this. The API has been enabled by default since Chrome 151 shipped on July 28. A soft navigation requires both a visible URL change and a visible paint.

Most sites will see nothing. Depending on platform, between 66% and 75% of CrUX origins have soft navigations on fewer than 5% of page views. Only 7.0% of Android origins and 10.6% of desktop origins have soft navigations on more than half. The picture differs for the 1,000 biggest sites, where soft navigations exceed half of views for 35% of Android origins and 42% of desktop ones.

Sullivan’s guidance is blunt. A site that never rewrites page content and the address bar through JavaScript, skipping a full browser reload, should see its metrics hold steady.

For heavy SPA sites, the numbers move in a favorable direction. Among Android origins where 75% to 100% of views are soft navigations, good LCP shows up for 26% under the stable measure and for 87% under the experimental measure. Sullivan attributes the jump to follow-up clicks being faster on every kind of site, not to SPAs outperforming multi-page sites.

INP changes because it now scores responsiveness per page view, not by the single slowest interaction of a whole session. Per Chrome, Android pass rates for SPA-dominant sites about double, yet those sites still trail multi-page sites on INP. CLS now groups layout shifts by view as well. Desktop SPA-heavy sites reach parity with multi-page sites, and Android remains slightly behind.

Across all origins, the September dataset shows 55.9 percent passing Core Web Vitals without soft navigations and 59.5 percent with them.

Note what these gains represent. The improvement comes from a change in how the work is counted, not from any site getting faster. The source includes no independent evidence that the new scoring tracks user experience better than the old scoring did.

The ranking angle needs care. The source says nothing about ranking effects, so these experimental fields should not be treated as a ranking factor. Chrome targets May 2027 for ending the experimental label, depending on feedback received. Search Console is to take on the data afterward, according to Chrome. Under the Web Vitals lifecycle, a metric spends a minimum of six months in a pending stage prior to reaching stable. The post does not say which stage follows the experimental one.

Chrome also warns readers to “expect some shifts from month to month as the measurement improves.” The November CrUX release should add FCP counted per soft navigation, along with navigation types.

SEO and engineering teams running SPAs should open CrUX Vis, switch on the soft navigation toggle, and record which of the three metrics diverge from the stable numbers and by how much. Treat the results as a baseline for planning, then recheck after each monthly release before redirecting any engineering budget.

Search Engine Journal, October 13, 2026, reporting on Chrome’s announcement of experimental soft navigation metrics in CrUX.