KeenMerch
KeenMerch answers

The Product Page Speed Check You Can Run Before Your Morning Coffee

Short answer: Largest Contentful Paint, LCP, measures how long the largest visible element on the page takes to load, and Google calls 2.5 seconds or less good, 2.5 to 4 seconds needs improvement, and over 4 seconds poor. The score is judged at the 75th percentile of real page loads, on mobile and desktop separately. Run PageSpeed Insights on your top three product pages, read the 75th percentile, and fix the largest element first.

The number that decides the first impression

Largest Contentful Paint is the Core Web Vital that measures loading performance, and it is the one that decides whether a shopper sees your product or leaves. LCP is the time from when the page starts loading to when the largest visible element, usually the hero image or the product photo, has rendered. Google's thresholds are simple: 2.5 seconds or less is good, 2.5 to 4 seconds needs improvement, and over 4 seconds is poor. The other two vitals sit alongside it: Interaction to Next Paint at 200 milliseconds or less, and Cumulative Layout Shift at 0.1 or less.

The score that matters is not the one you see on your own machine. Google judges the 75th percentile of real page loads, on mobile and desktop separately, which means a quarter of your shoppers can be having a bad experience while your own test looks fine. The 2025 Web Almanac found only 48 percent of mobile sites passed all three Core Web Vitals, which means more than half of the stores your shoppers visit are failing at least one of them. The bar is not high. The bar is just rarely cleared.

The five-minute check

Open PageSpeed Insights, paste the URL of your best-selling product page, and read the field data, not the lab data. Field data is what real visitors experienced. Lab data is what a simulated visit on a fast connection experienced, and it flatters you. Read the 75th percentile LCP, then scroll to the element the report names as the largest contentful paint. That element is the job. If it is a hero image, the fix is usually compression, a modern format, and a width that matches the display size. If it is a video or a slider, the fix is usually to stop loading it before the product photo.

Run the same check on your second and third best sellers, and on the collection page that feeds them. The pattern across three pages tells you whether the problem is one image or the template itself. A template problem shows up on every page. An image problem shows up on one. Fix the template problem first, because it pays on every page at once.

What to fix first, and what to leave alone

Fix the largest element first, then the render-blocking scripts, then the layout shift. The largest element is usually the product photo, and the fix is usually unglamorous: a smaller file, a modern format, and a preload hint so the browser fetches it before anything else. Render-blocking scripts are usually the third-party apps, the chat widget, the review widget, the tracking pixels, and the fix is to defer the ones that do not need to run before the page paints. Layout shift is usually an image without dimensions, and the fix is to set the width and height so the page does not jump when the image arrives.

Leave alone the things that do not move the number. A new theme, a redesign, and a font change are all expensive ways to avoid compressing an image. Run the check, fix the largest element, and run the check again. If the 75th percentile moved, you fixed the right thing. If it did not, the problem is somewhere else, and the report will tell you where. For the recovery side of the same funnel, empty-box returns and the weight check that catches them covers the return fraud that eats the margin a faster page earns, and the return policy that cuts abuse without cutting sales explains the policy that pairs with it.

Frequently asked questions

What is Largest Contentful Paint?

LCP measures how long the largest visible element on a page, usually the hero image or product photo, takes to load. Google calls 2.5 seconds or less good, 2.5 to 4 seconds needs improvement, and over 4 seconds poor.

What is the 75th percentile in Core Web Vitals?

Google judges the 75th percentile of real page loads, on mobile and desktop separately. It means 75 percent of visits meet the threshold, and the slowest quarter does not. Your own test on a fast connection is not the score.

How do I check my product page speed?

Run PageSpeed Insights on your top three product pages and read the field data, which is what real visitors experienced. Find the element the report names as the largest contentful paint, and fix that element first.

What should I fix first for a slow product page?

The largest element, usually the product photo: compress it, use a modern format, and set the width to match the display. Then defer the third-party scripts that block rendering, then set image dimensions to stop layout shift.