Page Speed Test UI

Lab metrics are timed measurements such as first contentful paint taken in a simulated load. A performance score rolls those into one number. This UI shows that kind of report without calling the PageSpeed service.

Mobile always shows performance 68, accessibility 85, best practices 92, and SEO 100, with FCP 2.4 s, LCP 4.1 s, TBT 320 ms, and CLS 0.15. Desktop always shows performance 92, accessibility 95, best practices 96, and SEO 100, with FCP 0.8 s, LCP 1.2 s, TBT 40 ms, and CLS 0.02.

Performance is one of the values Page Speed Test UI puts on screen. A rolled-up score from lab metrics. Mobile is 68 and desktop is 92, every time. Do not tell a client their performance score is 68 because this tab said so.

Read Accessibility, best practices, SEO on its own before you mix it with the other rows. Lighthouse also scores these categories. Mobile is 85, 92, and 100. Desktop is 95, 96, and 100. A perfect SEO 100 here is not an audit of the page.

FCP and LCP answers a narrower question than the headline number. First contentful paint and largest contentful paint are paint timings. Mobile shows 2.4 s and 4.1 s. Desktop shows 0.8 s and 1.2 s. Those seconds were not measured on your URL.

Treat TBT and CLS as a label with a specific job. Total blocking time and cumulative layout shift describe jank and movement. Mobile shows 320 ms and 0.15. Desktop shows 40 ms and 0.02. Do not file a layout-shift bug from 0.15 on this screen.

The Tabs line is worth a full stop. Lab tools show a phone run and a desktop run. Both tabs are fixtures. Switching tabs does not retest the URL.

If you need a lab run against a real URL, use the SEO Audit Tool, which asks the server to include a PageSpeed request. This form only draws the layout. The URL is parsed so a bad string can be rejected. Parsing is not a test.

If you remember one sequence from Page Speed Test UI, remember the fields in the order they change a decision. Performance matters because A rolled-up score from lab metrics. In practice, Mobile is 68 and desktop is 92, every time. The mistake to avoid is this: Do not tell a client their performance score is 68 because this tab said so. Accessibility, best practices, SEO matters because Lighthouse also scores these categories. In practice, Mobile is 85, 92, and 100. Desktop is 95, 96, and 100. The mistake to avoid is this: A perfect SEO 100 here is not an audit of the page. FCP and LCP matters because First contentful paint and largest contentful paint are paint timings. In practice, Mobile shows 2.4 s and 4.1 s. Desktop shows 0.8 s and 1.2 s. The mistake to avoid is this: Those seconds were not measured on your URL. TBT and CLS matters because Total blocking time and cumulative layout shift describe jank and movement. In practice, Mobile shows 320 ms and 0.15. Desktop shows 40 ms and 0.02. The mistake to avoid is this: Do not file a layout-shift bug from 0.15 on this screen. Tabs matters because Lab tools show a phone run and a desktop run. In practice, Both tabs are fixtures. The mistake to avoid is this: Switching tabs does not retest the URL. After that, the checks are simple. You compared two URLs and saw the numbers stay put. You did not publish mobile 68 as a measurement. You did not treat SEO 100 as a pass. You left CLS 0.15 out of a bug ticket. A real Lighthouse or PageSpeed run is the source for any change you will make.

Keep Page Speed Test UI as this step only. When the job moves on, the Page Size Checker is the next page: Page Size Checker asks for a URL and then shows a total kilobyte figure plus a split across HTML, images, scripts, and CSS. The Mobile-Friendly Checker covers a different piece of the same work: Mobile-Friendly Checker asks for a URL and then shows a pass or fail plus a short list of checks.