Chrome Revolutionizes Web Performance Measurement: Introducing Experimental Core Web Vitals for Soft Navigations

In a landmark shift for the web development ecosystem, Google’s Chrome team has officially introduced experimental versions of the three Core Web Vitals (CWV)—Largest Contentful Paint (LCP), Interaction to Next Paint (INP), and Cumulative Layout Shift (CLS)—within the Chrome User Experience Report (CrUX).

For years, Single-Page Applications (SPAs) have operated in a measurement vacuum. Because traditional monitoring tools were built for the classic "multi-page" web, SPAs—which update content dynamically without a full browser reload—often saw their subsequent user interactions unfairly attributed to the initial landing URL. This often resulted in skewed, punitive performance data. By integrating the Soft Navigations API into the CrUX dataset, Google is finally providing developers with a granular view of how modern, dynamic web applications actually perform in the wild.

The Chronology of the Update

The journey toward this release began in earnest with the Soft Navigations API, which became enabled by default in Chrome version 151 on July 28. This API provides the technical scaffolding required to define a "soft navigation": a user action that triggers both a visible change to the URL and a meaningful paint of new content.

The integration into CrUX was confirmed in the October 13 release. As of this date, the new experimental metrics are available across the entire CrUX ecosystem, including:

• BigQuery: Enabling deep, custom data analysis.
• CrUX API: Allowing developers to programmatically fetch data.
• History API: Providing a look back at how these metrics evolve.
• CrUX Vis: A user-friendly visualization tool that now includes a toggle to compare "stable" (traditional) vs. "experimental" (soft-nav-inclusive) metrics.

Google has established a long-term roadmap for this transition. The metrics will remain in an "experimental" state until May 2027. During this window, the Chrome team will be collecting developer feedback to refine the measurement logic. Only after these metrics graduate from the experimental phase will they be adopted by Google Search Console, which ultimately dictates how performance data influences SEO and ranking signals.

Understanding the "Soft Navigation" Mechanic

To appreciate why this change is transformative, one must understand how performance was previously measured. Before this update, CrUX only monitored the "hard load" of a page. If a user navigated through an SPA—clicking buttons, triggering AJAX requests, and updating the DOM—all subsequent layout shifts and interaction delays were grouped into the initial page load’s metrics. This created a "long-tail" effect where the initial page load bore the brunt of all errors that occurred throughout a user’s entire session.

The new experimental fields—experimental_largest_contentful_paint_with_softnavs, and their counterparts for INP and CLS—change the fundamental unit of measurement. Now, when a soft navigation is detected, it is tracked as an individual event. The metrics are associated with the specific destination URL, subject to standard eligibility criteria, providing a much more accurate reflection of the user experience at every step of the journey.

Supporting Data: Impact and Reach

Google’s internal analysis of the CrUX dataset reveals that this update will not affect every site equally. In fact, for the vast majority of the web, the change will be negligible.

The Landscape of Usage

Data shows that approximately 70% of origins tracked in CrUX exhibit little to no soft navigation traffic. For these sites, which rely on traditional page reloads, the experimental metrics will mirror the stable ones. However, the story shifts dramatically when looking at high-traffic, modern web applications:

• General Web: Only 7.0% of origins on Android and 10.6% on desktop rely on soft navigations for more than half of their page views.
• The Top 1,000: The reliance on soft navigations is significantly higher among the most popular sites, with 35% on Android and 42% on desktop using them for over 50% of their traffic.

Performance Gains for SPAs

Perhaps the most striking revelation from this release is the improvement in scores for SPA-heavy sites. For Android origins where 75% to 100% of traffic consists of soft navigations, the share of sites achieving "Good" LCP scores jumped from 26% (stable) to 87% (experimental).

This does not necessarily mean SPAs are faster than Multi-Page Applications (MPAs); rather, it acknowledges that "follow-up clicks" are inherently faster than initial document loads. By isolating these clicks, the metrics finally align the perceived performance of an SPA with that of an MPA, preventing the "first-load bias" that previously dragged down SPA scores.

Official Responses and Perspectives

Annie Sullivan, a key voice from the Chrome team, has been instrumental in clarifying the intent behind these changes. She emphasizes that developers should not interpret these improvements as an inherent superiority of the SPA architecture. Instead, it is a correction of a measurement bias.

"If your site doesn’t use JavaScript to dynamically swap out page content and update the URL without a full browser reload, you shouldn’t expect to see any changes in your metrics," Sullivan noted. She further explained that for sites with heavy soft-nav traffic, the improvement in INP—which now measures responsiveness per page view rather than the slowest interaction in an entire session—is a critical step toward fairness. By grouping layout shifts by "view" rather than "session," the new CLS metrics also bring SPA-heavy desktop sites into parity with traditional sites, narrowing the gap that previously existed.

Strategic Implications for Web Developers

For organizations heavily invested in React, Vue, Angular, or other SPA frameworks, this is a pivotal moment. The implications of this update are multi-faceted:

1. Re-evaluating Performance Audits

Developers can now use the "Include soft navigations" toggle in CrUX Vis to conduct a side-by-side audit. This provides an immediate visual representation of how the new methodology changes the site’s performance profile. If a site meets Google’s thresholds under the experimental metrics but fails under the stable ones, the team now has a data-backed argument that the site’s performance is better than the legacy metrics suggested.

2. Preparing for the 2027 Transition

While the metrics are currently experimental, the "pending" status in the Web Vitals lifecycle suggests they will eventually become the new standard. Google has signaled that the next checkpoint will arrive with the November CrUX release, which is expected to include experimental versions of First Contentful Paint (FCP) and detailed navigation types. Developers should use the next two years to optimize their SPA routing, as these metrics will eventually become a core component of the Google Search algorithm.

3. Caution and Continuous Monitoring

Google has explicitly warned that developers should interpret these numbers with caution. As the measurement logic matures, scores may fluctuate. The Chrome team encourages active participation in the Web Vitals feedback group and the Soft Navigations spec repository on GitHub. This is not a "set it and forget it" update; it is an invitation to engage in the standardization of how the modern web is measured.

Conclusion: A More Fair Web

The introduction of soft navigation metrics into the Core Web Vitals marks the end of an era where SPAs were judged by an outdated, page-reload-centric rubric. By acknowledging the unique way dynamic applications interact with the user, Google is ensuring that the metrics that drive business success—like search ranking and user experience optimization—are grounded in the reality of the user’s journey.

While we are currently in an experimental phase, the trajectory is clear. The web is moving toward a more nuanced understanding of performance, one that treats every user interaction as a distinct, meaningful event. For the developers building the next generation of digital experiences, the era of "fair metrics" has finally arrived. As we move toward the May 2027 target, the focus must remain on iteration, feedback, and, most importantly, the actual user experience that occurs between the clicks.