Free Audit · Performance
How slow is your site really?
Not a guess from your HTML. We open your pages in a real headless browser and measure what actually happens: how long the load takes, how much you ship, how many requests it costs, and what blocks the first paint.
- Load time measured in a real browser, not estimated
- Page weight and request count from the actual network log
- Render-blocking scripts and stylesheets, per page
- Oversized images, with the ones worth fixing first
Run your free audit
Two fields. Real browser measurement takes a minute or two — we email you when it is done.
Why five pages
Measured, not estimated.
Most free speed tools read your HTML and infer. We load each page in a real headless browser and take the numbers from the actual network log — which costs five to ten seconds a page, so we check five rather than thirty. That is not a limitation so much as an admission: the score is a median, and a handful of representative pages tells you the same thing thirty would, without keeping you waiting.
If you want every page measured, that is the full site audit — and if the headless browser is ever unavailable, the report tells you the figures are estimates rather than quietly presenting a guess as a measurement.
How it works
From URL to report.
Enter your URL
Your site address and your email. That is the whole form.
We load your pages
Each one opens in a real headless browser while we record the network and the timings.
Get your report
Median and p75 load time, weight and requests, plus what is blocking the first paint.
Common questions
Performance audit FAQs
Google's tool is excellent and you should use it. The difference is scope: it tests one URL at a time against lab and field data. This loads several of your pages in a real browser and gives you the median and the p75 across them, so you see whether a problem is one page or your whole template.
Because we measure rather than estimate. Every page opens in a real headless browser while we record the network log, which costs five to ten seconds each. The score is a median, so five representative pages tell you the same thing thirty would, without leaving you waiting.
The median is the typical visitor. The p75 is the unlucky quarter — the slower tail. If those two numbers are close, your site is consistent. If the p75 is much worse, something specific is dragging, and the per-page table will show you which page.
It means the headless browser was unavailable when your audit ran, so load times and weights were inferred from your HTML instead of measured. We tell you rather than quietly presenting a guess as a measurement. Re-run later for measured figures.
The measurements here are a desktop pass. Mobile performance is covered in the full site audit, which runs a separate mobile pass across more of your site.
Images, almost always. After that, render-blocking scripts in the head and third-party tags nobody remembers adding. None of it is glamorous and all of it is measurable.
Performance Audit
Want this faster, not just measured?
Most performance work is unglamorous and effective: fewer requests, smaller images, less blocking script. We do it.
A senior engineer reads it and replies within one business day.