WEBSITE PERFORMANCE
View Transitions API in production: UX data from three live sites.
Cross-document View Transitions shipped in Chrome 126 (June 2024) and reached broad support in early 2025. The API lets the browser animate between pages without a single-page-app. We shipped it on three client sites in Q4 2025 and Q1 2026. Here is what changed.
Site A. DTC skincare, category → product
Transition: product image morphs from the category grid into the product detail page. Roughly 90 lines of CSS and a view-transition-name on the image.
Metrics (30-day window, A/B on 50% of traffic):
Product-detail-page bounce rate: 42% → 38% (-4pp, p<0.01).
Add-to-cart rate per PDP view: +6% (p=0.02).
p75 LCP: flat (transition does not block LCP measurement).
Session revenue: +3.1% in the test cell (not quite significant at n=18k sessions).
Site B. B2B SaaS marketing site, pricing page
Transition: pricing tier cards slide smoothly when the user toggles monthly/annual or clicks between tiers. Same-document transition, not cross-document.
Metrics:
Time on pricing page: +22 seconds median.
Demo-request rate from pricing: flat.
Qualitative (session replay): noticeably more interactions with the toggle. Users played with it.
A "users engage more" outcome without a conversion lift. Ambiguous but probably harmless.
Site C. Editorial publisher, article → article
Transition: the article hero image and headline persist across next-article clicks within a series.
Metrics:
Articles-per-session: 1.8 → 2.3 (+28%).
p75 INP on transition: 220ms → 260ms (slight regression, attributable to the transition frame).
Ad impressions per session: +24%.
What we learned
View transitions feel native when done subtly. They feel gimmicky when done showily. Start with the smallest possible transition.
The API is safe to ship with progressive enhancement. Non-supporting browsers get a plain page navigation. There is no fallback to write.
INP regresses slightly during the transition frame. Not a reason to avoid it, but measure it.
The biggest gains were on editorial / content sites where "fluid reading experience" is the product. Ecommerce and B2B saw softer gains.
Sources: Chrome's View Transitions documentation; three client A/B tests, Oct 2025 to Feb 2026.