Lighthouse score calculator
Lighthouse’s 0–100 performance score isn’t a measurement; it’s a weighted blend of five lab metrics. Set the values and see where every point comes from.
Needs improvement
| Metric | Metric score | Points |
|---|---|---|
| FCP | 81/100 | 8.1 / 10 |
| SI | 78/100 | 7.8 / 10 |
| LCP | 61/100 | 15.3 / 25 |
| TBT | 63/100 | 18.9 / 30 |
| CLS | 94/100 | 23.6 / 25 |
Most points left on the table: Total Blocking Time (+11 possible).
How the score is built
Each metric is mapped to a 0–1 score on a log-normal curve defined by two control points taken from HTTP Archive data: the value that earns 0.9 (roughly the top 8% of sites) and the value that earns 0.5 (the median). The curves flatten at both ends, so shaving 200 ms off an already-fast LCP moves the score far less than the same saving on a slow one.
The weights in Lighthouse 10 and later are TBT 30%, LCP 25%, CLS 25%, FCP 10% and Speed Index 10%. Time to Interactive was removed in Lighthouse 10. Because TBT carries the most weight, JavaScript-heavy pages often score poorly even when they look fast.
| Metric | Weight | Mobile: 0.9 / 0.5 at | Desktop: 0.9 / 0.5 at |
|---|---|---|---|
| FCP | 10% | 1.80 s / 3.00 s | 0.93 s / 1.60 s |
| SI | 10% | 3.39 s / 5.80 s | 1.31 s / 2.30 s |
| LCP | 25% | 2.50 s / 4.00 s | 1.20 s / 2.40 s |
| TBT | 30% | 0.20 s / 0.60 s | 0.15 s / 0.35 s |
| CLS | 25% | 0.1 / 0.25 | 0.1 / 0.25 |
Source: Lighthouse scoring documentation (developer.chrome.com, “Lighthouse performance scoring”). The calculator reproduces the log-normal formula; results match Lighthouse within rounding.
What the score is (and isn’t) good for
Use it as a quick regression alarm in CI or after a deploy. Don’t use it as your target: Google ranks with Core Web Vitals from real users, and a page can score 70 in the lab while passing every field threshold. Run a real test to see both side by side.
Questions people ask
How is the Lighthouse performance score calculated?
Lighthouse scores five lab metrics on log-normal curves (0 to 1 each) and combines them with fixed weights: Total Blocking Time 30%, Largest Contentful Paint 25%, Cumulative Layout Shift 25%, First Contentful Paint 10% and Speed Index 10%. The weighted sum times 100 is the score.
What is a good Lighthouse score?
90–100 is shown as green (good), 50–89 as orange (needs improvement) and 0–49 as red (poor). A score of 100 is not required for good Core Web Vitals, and the score itself is not a ranking factor.
Why is my mobile score so much lower than desktop?
Mobile runs simulate a mid-range phone with a 4× slower CPU on a throttled 4G connection, and the mobile scoring curves are stricter in absolute terms. The same page commonly scores 30–40 points lower on mobile.
Is the Lighthouse score a Google ranking factor?
No. Google’s page experience signals use real-user Core Web Vitals from the Chrome UX Report, not the Lighthouse lab score.