# IndieSpeed > Check a page's mobile and desktop speed, Core Web Vitals, and the three fixes worth doing first. Mobile and desktop scores, real-user vitals, and the three fixes worth doing first. The canonical page is https://indiespeed.com/. ## Use the tool Open [IndieSpeed](https://indiespeed.com/) to complete the job. The questions below describe the supported input, result, and limits. Every answer names its source next to it. Source: [Google PageSpeed Insights](https://developers.google.com/speed/docs/insights/v5/about). ## Questions and limits ### What is a good PageSpeed score? The score is Lighthouse's performance score, from 0 to 100, measured in a lab run of the page. Google reads 90 and above as good, 50 to 89 as needing improvement, and below 50 as poor. The number is a weighted blend of five timings, and IndieSpeed lists them under each score: First Contentful Paint, Largest Contentful Paint, Total Blocking Time, Cumulative Layout Shift, and Speed Index. IndieSpeed runs the mobile and desktop tests through Google's [PageSpeed Insights](https://developers.google.com/speed/docs/insights/v5/about) and shows both scores side by side. ### What are Core Web Vitals? Core Web Vitals are three numbers Google collects from real Chrome visitors: Largest Contentful Paint, how long the main content takes to appear, Interaction to Next Paint, how long the page takes to respond to a tap or click, and Cumulative Layout Shift, how much the layout jumps. Good means at most 2.5 seconds, 200 milliseconds, and 0.1, measured at the 75th percentile of visits over 28 days. IndieSpeed shows them when Google's Chrome UX Report has enough visits to the page. When only the whole origin has enough, it labels the fallback and shows that instead. When neither has, it says not measured. A new site or a low-traffic page has no field data yet, and no lab run can stand in for it. ### Why is my mobile score lower than desktop? Lighthouse tests mobile on a simulated mid-range phone with a slow 4G connection and a throttled processor, and desktop on a fast machine with a fast connection. The same page takes longer to load and to run its scripts on the phone, so every timing is worse and the score drops. A page that scores 95 on desktop can score 60 on mobile without anything being wrong with the test. Most visits to most sites come from phones, so read the mobile score as the real one. ### Why does my PageSpeed score change every time? A Lighthouse score is one lab run, and a run depends on the network, the server, and the machine at that moment. Google says to expect variance, and a few points either way between runs is normal. A jump of twenty points usually means a script or a third party that behaves differently between loads. IndieSpeed keeps a successful measurement for six hours, so a repeat inside that window shows the same run. The result says when it was measured and when a fresh run is available. ### How do I improve my PageSpeed score? Start with the three fixes under the scores. IndieSpeed takes every improvement Lighthouse found on either device, ranks them by their score, their estimated savings, and their effect on the weighted timings, and keeps the top three. Each one carries plain advice, the devices it applies to, Google's estimated saving, and a link to Google's own guidance on it. The savings are estimates, not promised score gains. Make the first change, run the test again after the cache window, and take the next one. ### Does Google use PageSpeed score for ranking? Not the score itself. Google's ranking systems use the Core Web Vitals measured from real visitors, as part of the [page experience](https://developers.google.com/search/docs/appearance/page-experience) signals, and Google says content and relevance matter more. The Lighthouse score is a lab number that helps you find what to fix, not a number Google reads. So a green field section matters more than a green score. A page with no field data has not been measured for ranking either way. ## Promises - The core is free: no account, no card, no premium plan, no upgrade nag, no hidden paywall. - Private lookup activity is never published. - Cookieless analytics measures visits and incoming campaign attribution. It never receives entered values or file contents. No lookup history becomes public inventory. ## More - [Notes](https://indiespeed.com/blog.md): posts by the maker, newest first - [More tools](https://indiespeed.com/tools.md): the other tools by turushan - [security.txt](https://indiespeed.com/.well-known/security.txt): how to report a vulnerability Made by turushan, https://twitter.com/turushan.