What page size affects
A smaller, leaner HTML document generally loads and parses faster, which matters for both user experience and Core Web Vitals. This tool fetches a page's raw HTML and reports its size along with how many images, scripts, and stylesheets it references - a quick first signal of how heavy a page might be, before digging into a full performance audit.
Why this measures only the HTML document itself
This measures the HTML document itself - it does not download every image, script and stylesheet to total the full page weight. That's a deliberate scope: checking just the HTML document is fast and instant, while downloading every linked resource to total true page weight takes considerably longer and is better suited to a dedicated page-weight tool built for that specific job.
Reading resource counts alongside HTML size
A page with a modest HTML size but a very high script or stylesheet count is a signal worth investigating separately - each additional linked resource adds its own network request on top of whatever its file size turns out to be, so counting them gives you a sense of request overhead that raw HTML size alone doesn't capture.
When a large HTML document is actually worth investigating
There is no fixed threshold, but HTML over roughly 100-150 KB is often worth investigating for bloat (excessive inline styles, unminified markup, or embedded data like base64 images) - well-optimized pages are typically well under that, and finding the specific cause of HTML bloat usually points to a clear, fixable issue rather than a vague "the page is slow" problem.