Why does Lighthouse show different numbers than Google Search Console?
Google Lighthouse and Google Search Console provide different performance metrics because one offers a snapshot using synthetic tests, while the other shows aggregated real user data over time.
Because they measure performance at completely different scopes. One is a snapshot and the other is a trend, the numbers are not expected to match.
Google Lighthouse runs a synthetic test on a single page, at a single moment in time, using a simulated device and network. It assumes a first-time visitor, a cold cache, and no real user behavior. Its goal is to expose potential performance bottlenecks, not to describe how users actually experience the site.
Google Search Console, by contrast, reports aggregated performance from real users over time. Even when the data appears to be “for one page,” it represents many visits, across many devices and networks, summarized at the 75th percentile. It reflects what most users actually experienced, not a hypothetical scenario.