The Real Cost of a 1-Second LCP Delay on an Ecommerce Checkout

No study puts a precise dollar figure on one second of LCP delay on your checkout page specifically. LCP as a named metric is newer than most of the research linking page speed to revenue, and no public study isolates it at the exact moment a shopper is about to pay. What does exist, from a Google-commissioned study and independent conversion-rate research, all points the same direction: small amounts of load time move ecommerce conversion by whole percentage points, and it costs you fastest exactly where you'd expect, closest to the sale.

What the actual data says

• Deloitte and Google, "Milliseconds Make Millions." Across 37 leading retail and travel sites and over 30 million user sessions, a 0.1-second improvement across four speed metrics was tied to an 8.4% increase in retail conversion rate and a 9.2% increase in average spend. The funnel breakdown is the interesting part for checkout specifically: the same 0.1-second gain moved Product Listing to Product Detail progression by 3.2%, and Product Detail to Add to Basket by 9.1%, meaning the effect gets stronger, not weaker, the closer a shopper gets to buying.
• Portent's own conversion-rate tracking. Across roughly 100 million page views on 20 B2B and B2C sites, Portent found ecommerce conversion rates of 3.05% at a 1-second load time, dropping to 1.68% at 2 seconds, 1.12% at 3 seconds, and 0.67% at 4 seconds. A site loading in 1 second converts roughly 2.5 times better than one loading in 5.
• Google's own bounce and bloat research. Analysing real mobile session data, Google found that the probability of a mobile visitor bouncing rises 123% as load time goes from 1 to 10 seconds, and that conversion rate can fall by as much as 95% as a page's element count climbs from around 400 to 6,000. Bloat has a measured cost before you even get to raw speed.

We're leaving out the "100 milliseconds costs 1% of sales" line you may have seen attributed to Amazon. It traces back to a conference talk, not a published study, and it does not hold up to the sourcing standard the rest of this data does.

Why it matters most right at checkout

The average documented shopping cart abandonment rate, averaged across 50 separate studies by the Baymard Institute, is 70.22%. Most of that abandonment has nothing to do with speed: unexpected costs, forced account creation, a clunky form. But checkout is also the stage where a shopper has already decided to buy, which makes it the worst possible place to lose them to friction they never chose to tolerate. A slow product page costs you a browser who might come back. A slow checkout costs you a buyer who was already reaching for their card.

What usually causes slow LCP right at checkout

• Payment method iframes and SDKs. Card fields, wallet buttons, and buy-now-pay-later widgets are often third-party embeds loaded synchronously, and several competing for the same connection at once compounds the delay.
• Address autocomplete and tax or shipping calculation scripts. These frequently call an external API before rendering anything, and a slow or unthrottled call blocks the perceived load of the whole step.
• Trust badges and payment logos. Small individually, but rarely optimised, and often loaded from a third-party host rather than your own CDN.

The takeaway

Nobody has published the exact revenue cost of a 1-second LCP delay on a checkout page, and treat any number claiming otherwise with the same scepticism this site applies to every other unverified SEO stat. What the verified data shows is close enough: conversion is highly sensitive to speed, the sensitivity is strongest closest to the sale, and checkout is where a slow page costs you a buyer, not just a browser. If you don't know which script on your checkout is actually the drag, that's exactly what a Core Web Vitals audit traces to source.