Reviewed November 7, 2026

Speed Score vs Real LCP: The Number That Matters Is the User's

The speed score says 58, and the store owner panics. The customers load the page in two seconds, the checkout converts, and nothing feels slow. Both facts are true, because the score and the experience measure different things, and the score is the one that lies by omission.

The Shopify speed score is a Lighthouse lab score, run on a simulated connection, on a simulated device, at one moment. Largest Contentful Paint from real users, the field data Google actually ranks on, is measured on real devices, real networks, and real customers. A store can fail the lab and pass the field, and the field is the one that matters.

What the score actually measures

Lighthouse runs in a lab, with a throttled connection and a mid-tier phone simulation, and it scores the page against a fixed budget. Third-party scripts, app embeds, and tracking pixels all count against the score, even when they load after the page is usable. A store with ten apps will score worse than a store with none, whether or not the customer notices, because the score counts weight, not experience.

LCP, Largest Contentful Paint, measures when the largest visible element finishes loading, and the field version comes from the Chrome User Experience Report, real users on real devices. Google's threshold is 2.5 seconds for a good experience, measured at the 75th percentile, which means three quarters of real visits beat it. A store with an LCP of 2 seconds at the 75th percentile is fast, whatever the lab score says.

The two numbers diverge for a structural reason. The lab runs once, on a cold cache, on a simulated slow connection. Real users have warm caches, fast connections, and repeat visits. The lab is a worst case. The field is the average case. Optimizing for the lab means optimizing for a customer who does not exist.

When the score is worth chasing

The score is useful as a diagnostic, not as a target. A score below 40 with a bad field LCP means the page is genuinely slow, and the Lighthouse report will name the causes: an unoptimized hero image, a render-blocking script, a chat widget loading before the content. Fix those, and both numbers improve.

The score is not useful as a goal. Chasing a score of 90 usually means removing apps and features that earn money, to satisfy a simulation. The honest target is the field data: LCP under 2.5 seconds at the 75th percentile, Interaction to Next Paint under 200 milliseconds, and Cumulative Layout Shift under 0.1. Those are the numbers Google uses, and they are the numbers customers feel.

Check the field data in PageSpeed Insights, in the section labeled field data or discover what your real users are experiencing. If the field data is green and the lab score is orange, the store is fast and the score is noise. If both are red, the page needs work, and the lab report will say where.

The app trap

Every app adds weight, and the speed score counts it all, but the field data only counts what the user feels. An app that loads after interaction, or on the pages where it is used, barely touches LCP. An app that injects a script into the head of every page touches everything. The audit is not how many apps the store has. It is which apps load before the content, and whether they earn their weight.

Frequently asked questions

What is a good Shopify speed score?

The score is a lab number, not a target. A store with field LCP under 2.5 seconds at the 75th percentile is fast, even if the lab score is in the 50s or 60s.

What is LCP and why does it matter more than the score?

Largest Contentful Paint measures when the largest visible element loads for real users. Google ranks on field LCP from real users, not on the lab score, and customers feel LCP directly.

Why is my lab score low when my site feels fast?

The lab runs on a simulated slow connection with a cold cache, and it counts every third-party script. Real users have warm caches and faster connections, so the field data is usually better.

What LCP should I aim for?

Under 2.5 seconds at the 75th percentile of real users. That is Google's threshold for a good experience, and it is the number that correlates with conversion.