Performance
Does your site remain fast enough for its visitors? We measure page loading in a browser, Core Web Vitals, and the main factors that slow a site down: page weight, requests, code execution and third-party resources.
Try our method for free
One page. One pillar. A real report.
Enter your website and we’ll email you a report on Performance for your home page. Please allow at least 24 hours.
Our plans cover the 9 pillars across the pages of your whole site.
What we check
- How long the largest element on the page takes to appear, measured in a browser throttled to a phone and an ordinary network.
- Layout shifts during loading: content that moves under the finger at the exact moment you were about to tap a button.
- The time during which the page does not respond because it is still running code.
- The weight actually downloaded, the number of requests, and how they split between images, scripts, fonts and stylesheets.
- What blocks a fast paint, what is neither compressed nor cached, and images served at a size unrelated to the user’s screen.
- The share of that weight and that wait which belongs to third-party services rather than to your own site.
What we find on the sites we audit
Across about 30 websites we audited between August 14, 2026 and September 24, 2026:
- 35% take more than 4 seconds to display their main content
Each percentage covers the sites where the point could be checked. Many were audited because a defect showed up quickly, so these figures describe our audits, not websites in general.
See also: User experience & reliability, a real business website report.
On what basis
A finding is only worth something if it rests on an identifiable basis. Where one exists, we rely on a law, a standard or a specification, and for every point recorded the report names the page and gives the value we measured.
What it changes for you
Speed directly shapes your visitors’ experience. It is also one of the signals Google takes into account, without being decisive for ranking on its own.
A good one-off score is not enough: we look at what actually happens on the pages analyzed. Mobile conditions can reveal limits that are not visible on a fast desktop connection, so the report gives both mobile and desktop measurements for you to compare.
What this audit does not say
Our measurements cover a property of your site: its weight, its request count, its execution cost. A 20-megabyte site stays a 20-megabyte site, whatever connection its readers have.
They do not replace what your real visitors actually get, on devices and networks we do not control. Where real-usage data is available, we publish it alongside ours. If the two diverge, the gap is itself information: a heavy site served to a well-connected audience still excludes those who read it on a poor network.
Every limit of the examination is stated in the methodology note of your report. A point that cannot be verified is never presented as compliant.
The weight of this pillar
An overall score has to weigh what matters for your business: the 9 pillars do not carry the same weight depending on what your site is for. Here is what “performance” represents in the overall score, under each weighting. The report always states which one produced your score.
| Type of site | Share of the overall score |
|---|---|
| Type not determined | 14% |
| Business / brochure site | 18% |
| E-commerce site | 12% |
| Application / customer portal | 15% |
| Blog / media site | 16% |
| Public-sector site | 10% |
This pillar is examined alongside the other eight in every plan, whatever the number of pages. See pricing →
Guides for this pillar
Largest Contentful Paint: what delays it
LCP measures when a page's largest visible element appears. Google's thresholds, the four steps where time is lost, and what our audit measures.
Read the guide →Slow website: the causes in load order
A slow website rarely has a single cause. Here are the causes in the order a page runs into them, and what our audit measures at each step.
Read the guide →