Core Web Vitals Field vs Lab Data

SEO

Also: Field Data vs Lab Data · CrUX vs Lighthouse

Field dataReal visitors, real conditions

Lab dataSimulated test, one device

Google ranks onField data, not lab data

Use lab forDiagnosing before you ship

Quick definition

Field data and lab data are two different ways of measuring Core Web Vitals. Field data comes from real visitors using your site in real conditions, collected by Google's Chrome User Experience Report (CrUX). Lab data comes from a simulated test run once, on one device, in a controlled environment, usually via Lighthouse.

How it varies across Australia

Australian sites often show a wider gap between lab and field scores than global averages, mostly because of network distance to overseas servers. A page that passes comfortably in a lab test run from a fast connection can still fail the field data collected from Australian mobile users on slower networks.

Compare site performance across Australian industries →

Field data vs lab data

Field data(CrUX)

Real visitor data collected over a rolling 28-day period. This is what Google uses to rank you.

Source of truth for ranking

Lab data(Lighthouse)

A single simulated test on a fixed device and connection. Useful for diagnosis before you ship.

Not used for ranking

What it actually means

Think of lab data as a fitting room mirror and field data as a photo someone took of you in the wild. The mirror shows a controlled, flattering version. The photo shows what actually happened, with bad lighting, a different angle, and no do-overs.

Lab data comes from tools like Lighthouse and PageSpeed Insights running a single simulated visit on a fixed device and connection speed. It's repeatable, which makes it useful for testing whether a code change helped or hurt before you ship it. But it's not what Google uses to rank you.

Field data comes from the Chrome User Experience Report (CrUX), an aggregated feed of real visits from real Chrome users who've opted in to share performance data. It's messy, it includes visitors on old phones and patchy wifi, and it's a rolling 28-day average. This is the data that determines your Core Web Vitals pass or fail for search ranking purposes.

The gap between the two is where most confusion about page speed lives. A developer optimises against the lab score, ships it, and wonders why the site still fails in Search Console. The answer is almost always: the lab test didn't reflect the real conditions of the actual audience.

Lab data tells you why a page might be slow. Field data tells you whether it actually was, for the people who mattered.

How it shows up

The gap shows up most obviously in Search Console's Core Web Vitals report, where URLs are grouped by real-user field data into good, needs improvement, or poor. Compare that against a PageSpeed Insights lab run on the same URL and the scores frequently disagree, sometimes sharply.

It also shows up when a business reports 'we pass Core Web Vitals' based on an internal Lighthouse audit, only to find their conversion rate and bounce rate tell a different story once real user experience is accounted for. The lab test missed the slow mobile network, the older device, or the third-party script that only loads for certain visitors.

The Australian context

For Australian businesses hosting infrastructure overseas, this gap is often larger than teams expect. A lab test run from a data centre with a fast connection can look fine, while field data collected from real Australian mobile users on regional networks shows meaningfully worse loading speed. If your Search Console field data is failing but your lab test passes, server location and content delivery network setup are the first places to check.

Where people get this wrong

Reporting lab scores as if they were the ranking signal.Google uses field data from CrUX for ranking decisions, not the one-off lab test. A perfect Lighthouse score with poor field data still fails Core Web Vitals in the eyes of search.

Ignoring lab data entirely because it isn't the ranking signal.Lab data is still the fastest way to test a fix before it reaches real users. Skipping it means you find out whether a change worked only after it's live for a month.

Testing on a single device or location and assuming it represents everyone.Field data aggregates across every device, connection type and location that actually visits your site. A single lab run on a fast desktop tells you nothing about the mobile visitor on a weak connection who makes up most of your traffic.

Common questions

Which one does Google actually use to rank my site?

Field data, collected through the Chrome User Experience Report (CrUX). It's the rolling 28-day average of real visitor experience. Your lab test score from Lighthouse or PageSpeed Insights is a diagnostic tool, not the ranking input.

Why does my lab score pass but Search Console shows failures?

Lab tests run once, on one device and connection, usually a fast one. Field data reflects every real visit, including slower phones and patchy networks. The gap usually points to real-world conditions the lab test never simulated.

Should I ignore lab data if it doesn't affect ranking?

No. Lab data is the fastest way to test whether a code change improved or hurt performance before it reaches real users. Use it during development, then confirm the result in field data once it's live for a few weeks.

How long until field data reflects a fix I've shipped?

Field data in the Chrome User Experience Report is a rolling 28-day window, so improvements phase in gradually rather than appearing overnight. Expect two to four weeks before Search Console fully reflects a shipped fix.

Debrief

Get the next one

No spam. No fluff. Just the next article, straight to your inbox.

Keep exploring

About New Rebellion

New Rebellion is a marketing intelligence consultancy. We build tools, score Australian businesses on how their marketing actually performs, and publish Debrief every day. This dictionary is part of how we work in the open.

How we think →