Is Your Web App Failing Core Web Vitals? Chrome Changed the Test

Chrome now measures Core Web Vitals across in-app page changes. See if your single-page app is affected, what to check in CrUX, and what to do before May 2027.

If your website or web app swaps content without a full page reload, the way Google measures its speed is about to change. On October 13, 2026, Chrome published experimental Core Web Vitals that count "soft navigations", the in-app page changes common in single-page apps (SPAs).

Nothing changes in your official scores today. But Google plans to move Search Console onto this data later, so the numbers you see now are a preview of the numbers you will be judged on.

What Chrome actually changed

Until now, the Chrome UX Report (CrUX) treated an SPA visit as one long page. The first load was measured, and every route change after it was largely invisible. The new experimental series treats a recognised route change as another page view, so each one can contribute its own scores.

A soft navigation here means an action that produces both a visible URL change and a visible paint, without a full browser reload. The three new fields in CrUX are:

• `experimental_largest_contentful_paint_with_softnavs` for loading speed
• `experimental_interaction_to_next_paint_with_softnavs` for responsiveness, measured per view instead of across the whole session
• `experimental_cumulative_layout_shift_with_softnavs` for visual stability, segmented per view and URL

The data is in BigQuery, the CrUX API and the CrUX History API. Chrome says the stable scores stay unchanged while the experimental versions are evaluated, and it is aiming to graduate them in May 2027, subject to feedback. Search Console will switch to the new data once that happens.

Who this affects (and who can relax)

Chrome says about 70% of CrUX origins have little or no soft-navigation traffic. If your site is a classic multi-page site, such as most WordPress, Shopify or brochure sites, you should not expect any change. In the words of the announcement, sites without client-side route changes should see none.

The shift lands on a smaller group. Roughly 7–11% of origins get more than half their page views from soft navigations. Among the top 1,000 sites, 35% on Android and 42% on desktop fall in that bucket. If you run a React, Vue or Angular app, a web-based SaaS product, a dashboard, or a headless storefront, this is your update.

Your scores may go up. Don't read too much into it.

For origins where 75–100% of Android page views are soft navigations, Chrome reports the share of "good" LCP rising from 26% on the stable measure to 87% on the experimental one. Responsiveness (INP) roughly doubles its Android pass rate for these sites, and layout shift moves closer to multi-page sites.

Across the whole September dataset, according to Chrome's release notes as summarised by CSS Wizardry, the share of origins passing all three vitals rose from 55.9% to 59.5%.

Before you celebrate, read the reason. Chrome attributes the LCP gain to follow-up views reusing code and assets that were already loaded, not to SPAs being inherently fast. The first visit, which is the one a new customer from search or an ad actually has, is still the hard part. A better experimental score does not mean a faster first impression.

Where problems can still hide

Chrome is clear that heavy main-thread JavaScript during route transitions can still hurt INP. Per-view measurement also changes the shape of the problem. A slow interaction on one screen is no longer averaged against the rest of a long session, so a single clunky checkout step or filter panel can show up on its own URL.

Expect some month-to-month movement too. Chrome warns the metrics are experimental and scores may shift as measurement improves, and the stable and experimental series cover different populations, so don't mix them on one chart.

How to check your own site this week

1. Find out whether you are affected. Look at your analytics: what share of page views come from in-app route changes? If it is low, stop here.
2. Open CrUX Vis and toggle on soft navigations for your origin, or query the three experimental fields through the CrUX API or BigQuery.
3. Compare stable and experimental scores side by side and note the gap per metric.
4. Check key destination URLs, such as pricing, product and checkout pages. They get their own soft-navigation data if they meet the normal CrUX eligibility rules.
5. Add soft-navigation measurement to your own real user monitoring, following Chrome's soft navigations documentation, so you are not relying only on a monthly public dataset.

What to do next

• Don't rewrite your front end because of an experimental metric. Nothing is scored differently in Search Console until after May 2027.
• Do fix what the new view exposes: slow route transitions, heavy scripts on interaction, and layout shifts when views swap in.
• Keep working on the first load. It is still where paid and organic visitors form their opinion, and where conversion rate optimisation usually pays back fastest.
• Watch the November CrUX release, which Chrome says will add experimental navigation-type and FCP metrics.

If your app is built on a JavaScript framework, speed work often sits across design, code and tracking at once. Our technical SEO team audits how Core Web Vitals and crawlability interact, our web development team fixes the slow routes and heavy scripts, and our web analytics setup service puts proper measurement in place so you can see real user data on every view.

Not sure whether your site counts as a single-page app, or what your scores look like under the new test? We are happy to look. You can get a free quote and we will tell you plainly whether there is anything to fix.

Sources

• Chrome UX Report: experimental soft navigation Core Web Vitals (web.dev)
• Web-Perf Wednesday 013: CrUX adds soft navigation Core Web Vitals (CSS Wizardry)
• Soft Navigations API (Chrome for Developers)
• CrUX Vis