Performance Lab
Scores from sites I have built, measured on the live sites with Google’s PageSpeed Insights. Nothing here is estimated or illustrative. If a category has not been tested, it shows a dash instead of a number.
What has been recorded so far: mobile Performance only. Accessibility, Best Practices, SEO, and desktop results have not been recorded for these sites yet and will be added with their test dates.
Dashboard
Filter by site, device, or category. Each row links to the case study and to a fresh PageSpeed Insights test you can run yourself.
| Site | Device | Performance | Accessibility | Best Practices | SEO | Tested | Report |
|---|---|---|---|---|---|---|---|
| Mountain InnHospitality | Mobile | 92 | — | — | — | median of 6 runs | Run fresh test for Mountain Inn (opens in new tab)original not saved |
| Deer Creek StoneStone quarry | Mobile | 95 | — | — | — | median of 6 runs | Run fresh test for Deer Creek Stone (opens in new tab)original not saved |
| Celebration Event & VenueEvent venue | Mobile | 98 | — | — | — | median of 6 runs | Run fresh test for Celebration Event & Venue (opens in new tab)original not saved |
| Referees Sports Bar & GrillRestaurant & bar | Mobile | 98 | — | — | — | median of 6 runs | Run fresh test for Referees Sports Bar & Grill (opens in new tab)original not saved |
| Gsudz DetailingAuto detailing | Mobile | 98 | — | — | — | median of 6 runs | Run fresh test for Gsudz Detailing (opens in new tab)original not saved |
| Lias’ Cleaning Co.Home cleaning | Mobile | 97 | — | — | — | median of 6 runs | Run fresh test for Lias’ Cleaning Co. (opens in new tab)original not saved |
| Makeover by MariaLash studio | Mobile | 95 | — | — | — | median of 3 runs | Run fresh test for Makeover by Maria (opens in new tab)original not saved |
| Josh Jordan for SheriffPolitical campaign | Mobile | 98 | — | — | — | median of 6 runs | Run fresh test for Josh Jordan for Sheriff (opens in new tab)original not saved |
| NightShift HotelsSaaS + website | Mobile | 99 | — | — | — | median of 6 runs | Run fresh test for NightShift Hotels (opens in new tab)original not saved |
Reading the scores
Lighthouse is a useful, repeatable check. It is not a verdict on a whole website. Here is what each of the four scores actually covers.
Measures: how quickly the page shows content, becomes usable, and stays visually stable while loading, in a simulated test on one device and connection.
Does not prove: how fast the site is for every real visitor. That depends on their phone, network, and location.
Measures: a set of automated checks, such as text contrast, image alt text, form labels, and button names.
Does not prove: that a site is fully accessible or legally compliant. Automated tools catch only part of the problems; keyboard and screen-reader testing by a person is still needed.
Measures: selected technical checks, such as HTTPS, browser console errors, image aspect ratios, and deprecated APIs.
Does not prove: that the site is secure. It is not a security audit.
Measures: basic technical SEO checks, such as a title, a meta description, crawlable links, a valid robots.txt, and a mobile viewport.
Does not prove: that a page will be indexed, rank well, or be cited by AI search tools. A 100 means the basics are in place, nothing more.
Mobile vs. desktop
Lighthouse’s mobile test simulates a mid-range phone on a slower network, so mobile scores are usually lower than desktop scores for the same page. Most local-business visitors arrive on a phone, which is why the mobile number is the one I record first. Desktop still matters for office buyers, planners, and contractors who research at a desk.
Lab vs. field data
Every score on this page is lab data: one controlled test, repeated and summarized. Field data comes from real Chrome users visiting the site over the previous 28 days (the Chrome UX Report), and it only exists once a site has enough traffic. Small local sites often have no field data at all, which is why lab tests are the practical baseline. Where field data exists, it is the better measure of what visitors actually experience.
Method
Automated testing
The PageSpeed Insights API returns all four categories for mobile and desktop and is free to use with a Google Cloud API key, within Google’s published daily quotas. A scheduled job could re-test each site monthly and add dated records here automatically. It is not switched on yet: it needs an API key stored as a server secret, and every automated result would still be saved with its date and report link.
The website review tool uses the same API to test any public site on request.
Run the free review. You get the automated checks right away, clearly marked as automated.