Back to glossary

Term

Field Data vs. Lab Data

Field data are real performance values from actual users (CrUX, RUM). Lab data are synthetic measurements under controlled conditions (Lighthouse, WebPageTest). Google ranks by field data — lab data only hints at trends.

Field Data vs. Lab Data — explained in more detail

Field data comes from real user sessions: devices, networks, and conditions are as diverse as the site’s audience. Sources: CrUX (Google) for Chrome users, your own RUM via Cloudflare, Sentry, SpeedCurve, etc. Upside: shows lived performance. Downside: aggregates many factors, harder to debug causally.

Lab data comes from synthetic tests: a defined device, throttled network, cold-cache setup. Sources: Lighthouse (local/CI), WebPageTest, GTmetrix. Upside: reproducible, great for regression testing and debugging. Downside: doesn’t capture real-world variance — a site that performs well in the lab can fail in the field.

Example / In practice

A Lighthouse audit reports LCP at 1.5 s, but CrUX shows 4.2 s at the 75th percentile — typical when the lab profile simulates a modern mobile device while real users are on low-end Android. Adjusting the lab setup or supplementing with field telemetry yields more realistic targets.

Distinction from similar terms

RUM (Real User Monitoring) is a field-data method. Synthetic monitoring is lab data run continuously (e.g. daily WebPageTest). Google scores only field data for Core Web Vitals rankings — lab data is for diagnostics.

Entdecke mehr