We Checked 1 Million Online Stores. The Biggest Ones Pass Core Web Vitals Least Often, and Even Shopify's Own Store Falls Short.

Shopify built a company on the promise of fast, frictionless online stores. In Google's own field data, Shopify's own storefront does not pass Core Web Vitals, and neither do most of the biggest stores on the web. Rank online stores by popularity and the intuitive story reverses: just 36% of the top stores pass on mobile, versus about 65% of the small ones. The best-resourced retailers on the internet pass least often.

For a store, this is not an abstract score. Speed is conversion. Google's research ties slower mobile pages to sharply higher bounce rates, and retail experiments have repeatedly shown revenue moving with fractions of a second. When Nespresso, Tanishq or Shopify's own shop takes close to a second to respond after a tap, the shopper feels it at the exact moment they are trying to buy, and a share of them leave. That the flagships carry the heaviest experiences is what makes this surprising and worth reading.

The numbers come from the Chrome User Experience Report, Google's public record of what real Chrome users experienced on real devices and networks, and the same field signals Google's ranking systems use. Measured across roughly 1.07 million stores, the pattern is clean, monotonic and entirely about one metric.

We took about 1.07 million origins from the Chrome User Experience Report that run a recognized e-commerce platform (Shopify, WooCommerce, Magento and Adobe Commerce, BigCommerce, PrestaShop, Shopware, Salesforce Commerce Cloud and peers), ranked them by CrUX popularity, and grouped them into tiers, all measured on mobile at the 75th percentile. Store detection is fingerprint-based.

A store passes Core Web Vitals only when all three metrics land in the good range at the 75th percentile: Largest Contentful Paint under 2.5 seconds for loading, Interaction to Next Paint under 200 milliseconds for responsiveness, and Cumulative Layout Shift under 0.1 for visual stability. Every figure below is a real reading in Google's public dataset, and two-thirds of ordinary small stores already clear the bar, so it is plainly achievable.

There is an intuitive story about e-commerce speed: the giants, with their engineering teams, CDNs and performance budgets, should be fast, and the small independent stores on stock themes should be the stragglers. Google's field data reads the other way, cleanly and at scale. We ranked about 1.07 million measured stores by popularity and joined each to its real-world Core Web Vitals.

Among the roughly 140 most-visited stores on the web, 36.0% pass on mobile. Step down to the top 100,000 and it rises to 50.8%. The top million reaches 66.3%. And the long tail of ordinary small stores, over 980,000 of them, passes at 64.7%. The curve is monotonic where it matters: the more popular the store, the less often it delivers a good mobile experience. The best-resourced retailers on the internet pass least often.

The reason is not loading. Good LCP rates barely move across the tiers, with every group sitting in the low to mid 70s. What moves is responsiveness. Interaction to Next Paint, the metric that captures how quickly a page reacts when you tap, is good on 94.6% of long-tail stores and just 45.3% of the biggest ones. The median small store answers a tap in 100 milliseconds. The median top-10k store takes 225, past the 200 ms passing threshold. More than half of the web's flagship stores now sit above Google's bar for responsiveness on mobile.

Scale brings features, and features are JavaScript. The flagship store has search-as-you-type, personalized recommendations, a live mini-cart, promo carousels, a cookie banner, a tag manager loading a dozen marketing scripts, an A/B testing framework and a chat widget. All of it competes for the main thread the instant a shopper tries to do something. Each of those choices exists to move revenue, and each asks something of the main thread. The small store on a stock theme passes more often largely because it is asking for less. In the heaviest cases the delay is severe: several flagship stores post interaction readings of 400 to 675 milliseconds, roughly a second of visible pause after a tap while the main thread finishes its work.

Layout stability tells a quieter version of the same story. Good CLS rates soften as stores get bigger, from 93% in the tail to 77% at the top, so the biggest stores are both slower to react and more prone to shifting while they load. The full tier table lays the pattern out: loading flat across the board, responsiveness and stability both eroding as stores get bigger. It is an interactivity story from top to bottom.

Mobile, 75th percentile. Loading (LCP) is flat across tiers. Responsiveness (INP) and stability (CLS) both move as stores get bigger. The pattern is an interactivity story.

If you run a store, the fix is unusually well-targeted, because the shortfall lives in one metric. Good LCP holds in the low-to-mid 70s across every tier and good CLS only softens to 77% at the top, so loading and layout are rarely what fails you. INP is. Good INP falls from 94.6% in the long tail to 45.3% among the top 10,000, and the median top store answers a tap in 225 milliseconds against the 100 milliseconds a median small store manages. That is a main-thread JavaScript problem, and it is the number to chase.

The work is to audit what runs on the main thread the instant a shopper taps. Search-as-you-type, personalized recommendations, the live mini-cart, promo carousels, the cookie banner, the tag manager, the A/B test framework and the chat widget all compete for the same thread at the moment of interaction, and the heaviest stores show it: Tanishq answers in 675 milliseconds and fon.bet in 600. Defer or lazy-load anything not needed for the first interaction, break up the long tasks, and measure INP in the field rather than trusting a desktop lab score.

Shopify's own storefront failing at 300 milliseconds is the useful tell. It is not a gap in expertise, it is what ships when a store loads a lot of JavaScript to a phone, and two-thirds of ordinary small stores clear the bar precisely by staying light. Every conversion feature you add lands in the responsiveness Google measures and shoppers feel, so budget for it, keep the main thread the scarce resource it is, and treat a passing INP as part of the checkout.