How to use the page size checker

1Enter the page address and press Check page size.
2Read HTML document: Transferred is what crossed the network, Uncompressed HTML is the decoded size.
3Press Measure total page weight to include images, scripts, CSS and fonts (via PageSpeed Insights, 20–60 seconds).
4Compare the totals and fix the biggest part first.

Example: how the numbers relate (illustrative values)

Input

Transferred: 18,412 bytes with content-encoding: br
Decoded HTML: 96,830 bytes

Output

Transferred: 18 KB (br compressed)
Uncompressed HTML: 97 KB
Saved by compression: 81%

Saved = 1 − transferred ÷ uncompressed. Sizes use 1 KB = 1,000 bytes.

HTML size vs total page weight

The HTML document is the first file the browser downloads; everything else (images, CSS, JavaScript, fonts, video) is listed inside it and fetched afterwards. On most sites the HTML is a small part of the total; images and JavaScript dominate. This checker measures the HTML directly from our server and counts the scripts, stylesheets, images and iframes it references. For the full total, it runs a Lighthouse test through PageSpeed Insights, which loads the page in a real browser and adds up every request.

Breakdown by resource type

The HTML check lists how many external scripts, stylesheets, images and iframes the page references, which hints where the weight is. A per-type breakdown in bytes, and the largest individual files, are in the full PageSpeed Insights report (the "Avoid enormous network payloads" audit) and in your browser's developer tools: open the Network tab, reload, and sort by size.

Compressed vs uncompressed bytes

Servers can compress text files (HTML, CSS, JavaScript, SVG, JSON) with gzip or Brotli before sending them; the browser unpacks them. Transferred size affects download time; uncompressed size affects how much the browser has to parse. HTML usually compresses very well, so if content-encoding is missing, turning compression on is one of the cheapest speed wins there is. Images and video are already compressed and don't shrink further.

Largest resources to fix first

Images: usually the heaviest. Resize to the displayed size and compress; the image compressor converts to WebP.
JavaScript: large bundles and third-party tags; remove what you don't use and minify the rest.
Fonts: subset them and limit weights and styles.
CSS: remove unused rules and minify with the CSS minifier.
HTML: inline data and huge menus inflate it; minify with the HTML minifier.

What is a reasonable page size?

There's no official limit for visitors. Lighthouse flags pages whose total transfer exceeds about 5,000 KiB ("Avoid enormous network payloads") and treats lighter pages as better. For HTML specifically, Googlebot reads only the first 15 MB of an HTML file, so content past that point isn't indexed; real pages rarely come close. A lighter page loads faster on slow mobile connections and costs visitors on metered plans less.

What happens to the address I enter?

This tool sends the address you enter to Google PageSpeed Insights to measure the total weight of everything the page loads, only when you run the optional full-page test (the HTML check itself runs on our server). Their privacy policy applies to that request. Some checks also request the page from our server, which doesn't store what it fetches. We don't store the results.

Sources

Questions about the page size checker

What is a good page size?

As small as the content allows. Under about 1–2 MB total transfer loads comfortably on mobile; Lighthouse starts warning above roughly 5 MB.

Does Googlebot have an HTML size limit?

Yes: Googlebot fetches the first 15 MB of an HTML file (or other supported text file). Anything after that isn't used for indexing. Each image or script referenced is fetched separately with its own limit.

Why is the transferred size smaller?

Because the server compressed the file with gzip or Brotli. The browser downloads the smaller version and decompresses it.

Page Size Checker