See What Sticks.

PageSpeed Insights

Every other tool tests one page at a time. Run up to 25 simultaneous tests via the official Google PSI API — multiple URLs, multiple runs, all at once.

Batch. Benchmark. Know what's real.

25 Tests. One Click.

World First Browser OS

Launch from Quick Launch

150+ Tools

One Score Is Noise.
Five Runs Is Signal.

PSI scores fluctuate. Run five times and see the spread — not just the lucky hit.

You run PageSpeed Insights. You get 74. You run it again. You get 81. Which one is real?

Neither. And both. Google’s PSI API returns variance between runs because server load, CDN cold caches, and network conditions all shift. A single score is a snapshot — not a verdict.

Run multiple passes on the same URL and see the full spread: LCP, FID, CLS, FCP, TTFB, and overall Performance Score for every run. Toggle outlier runs off. Re-run a single test without clearing the rest. See the stable baseline, not the best-day score you happened to catch. Benchmark first. The technical SEO tools for complete site checks handle links, schema, and sitemaps next.

High variance between runs signals something real: inconsistent server response times, CDN cache misses, or render-blocking resources that only sometimes fire. Low variance means your numbers are trustworthy. Now you know the difference.

The Official Google API.
Basically Free.

No proxies. No third-party servers. Calls go directly to Google’s PageSpeed Insights API.

This tool calls Google’s official PageSpeed Insights API directly from your browser. The same API powering Google’s own PSI interface. The same data. The same scores.

Google’s API rate limits are generous enough that running 25 tests in a session costs nothing for most use cases. No signup required beyond your own API key if you need higher volume. No data leaves your browser except the URL you’re testing — sent directly to Google, the same as using their website.

Lighthouse vs PSI: Lighthouse is better for deep local audits — it runs in your browser or CLI and catches issues in a controlled environment. PSI is better for quick real-world numbers you can compare across multiple runs or multiple pages. For variance testing and competitive benchmarking, PSI is the right tool. That’s what this is built for.

Test Multiple Pages.
At the Same Time.

Homepage, blog, product pages — run all of them in a single session.

Launch from Quick Launch (Ctrl+Shift+Space, type “pagespeed”) or your assigned Quick Action shortcut. Paste any URL. Set your sample count. Results appear per run as each audit completes.

Core Web Vitals per run: LCP, FID, CLS, FCP, TTFB, and the overall Performance Score — displayed for every individual run so you can see the spread at a glance.

Per-run toggles. Suspect an outlier? Toggle that run off and see how the remaining results compare. Re-run individual tests without clearing the rest — something PSI’s own interface can’t do.

Error badges. When a run fails — API timeout, network error, invalid URL — the row shows an error badge with the reason. No silent failures. No empty rows with no explanation.

Reduced concurrency runs one audit at a time, waits for completion, then fires the next. Slower in wall-clock time but every run completes cleanly. For most use cases: keep concurrency reduced.