What Is Google's 2 MB HTML Crawl Limit? Complete Guide

How the 2 MB Limit Actually Works

Googlebot fetches a URL, and as the HTML streams in it counts bytes. The moment the running total reaches 2 MB, it stops reading the file. Per Google's official Googlebot documentation, it "only sends the already downloaded part of the file for indexing consideration." Everything after the 2 MB mark — including links Google would otherwise follow — is simply not seen.

Two details matter most. First, the measurement is on uncompressed data: a 500 KB gzipped response that decompresses to 3 MB of HTML is over the limit, because Google counts the 3 MB. Second, the limit is per file. The HTML document has its own 2 MB budget; each referenced image, stylesheet, and script is a separate fetch bound by the same per-file limit. So a page can be 8 MB in total transfer weight and still be perfectly fine for crawling, as long as the HTML file alone is under 2 MB.

What Counts Toward the 2 MB — and What Doesn't

The distinction that trips people up is "HTML document" versus "page." Only bytes that live inside the HTML file count.

This is why modern JavaScript frameworks are the usual culprit. A server-rendered React or Next.js page often inlines its serialized component data and hydration payload directly into <script> tags. That inlined data lives inside the HTML file, so it counts — and on data-heavy pages it can add hundreds of kilobytes to megabytes.

When Does a Page Actually Reach 2 MB?

Ordinary content pages almost never hit the limit. These specific patterns do:

1. Large inlined framework state. A server-rendered app that serializes a big dataset (a full product catalog, a long feed, a large table of rows) into the initial HTML.
2. Base64 images in the HTML. Embedding images as data: URIs instead of linking them can push a single document over 2 MB fast.
3. Auto-generated mega-pages. Programmatically built pages that dump thousands of rows, comments, or list items into one HTML response with no pagination.
4. Endless comment or thread pages. Forum and social pages that render every reply inline instead of paginating or lazy-loading.
5. Accidental duplication. Templates that inline the same large block (a full menu, a giant JSON-LD graph) multiple times per page.

How to Measure Your HTML Document Size

You want the uncompressed size of the HTML file only. Three reliable ways:

• Chrome DevTools. Open the Network tab, reload the page, click the top document request, and read the response size. Use the "Size" column's uncompressed value (hover to see transferred vs. resource size) — the resource size is what Google counts.
• View source. Right-click and "View Page Source," then save the file and check its byte size on disk. This is the raw HTML document, exactly what Googlebot downloads.
• Command line. Run curl -s https://example.com/page | wc -c to print the byte count of the raw HTML response.
• Greadme Deep Scan. Greadme measures the raw, uncompressed HTML document (before any CSS/JS stripping) and flags any page approaching or exceeding the 2 MB limit as part of a full audit — no manual measuring required.

A common mistake is measuring the rendered DOM in the Elements panel instead of the raw response. Google's limit is on the file it downloads, so always measure the initial HTML response, not the live DOM after JavaScript has run.

How to Fix an Oversized HTML Document

1. Move inlined data out of the HTML. Fetch large datasets on the client or from an API endpoint instead of serializing them into the initial HTML. Keep only the content that must be indexed in the document.
2. Paginate or lazy-load long lists. Render the first screenful server-side and load the rest on scroll or via separate paginated URLs that are individually crawlable.
3. Stop embedding base64 images. Link images with <img src> so they become separate fetches instead of inflating the HTML file.
4. Trim duplicated markup. Remove repeated inline blocks and oversized inline JSON-LD; keep structured data to what the page actually needs.
5. Put the important content first. Because Google reads top-down and stops at 2 MB, make sure your primary content, links, and structured data appear well before any large inline payload.

2 MB vs 15 MB: Two Different Limits

If you have read older SEO advice, you may remember a 15 MB figure from Google's June 2022 "Googlebot and the 15 MB thing" post. Both numbers are real, but they cover different scopes. The 2 MB limit is the one Googlebot applies to HTML and text files for Search indexing — the figure that decides whether your content gets indexed. The 15 MB limit is the default file size for Google's broader crawling infrastructure.

In early 2026, Google updated its documentation to state the 2 MB figure for HTML and text files explicitly. Google described this as a documentation clarification of the existing per-file-type limits, not a reduction in how much Googlebot crawls. For SEO, use 2 MB as your working target for the HTML document, and keep it lean either way.

FAQ

Does the 2 MB limit include images, CSS, and JavaScript?

No. It applies to the HTML document only. Images, stylesheets, and external scripts are fetched as separate files, each with its own 2 MB limit. Your total page weight can be much larger than 2 MB and still be fine.

Is the 2 MB measured before or after compression?

Before — it is the uncompressed size. A response that is 400 KB over the wire with gzip but decompresses to 2.5 MB of HTML is over the limit, because Google counts the decompressed 2.5 MB.

What happens to content after the first 2 MB?

Googlebot stops reading at 2 MB and only sends the already-downloaded portion for indexing. Any text, links, or structured data below the cutoff is never processed, so it cannot be indexed or followed.

My page is a few hundred kilobytes — do I need to worry?

No. A typical HTML document is 50–300 KB, well under the limit. This check only matters for pages that inline very large amounts of data or markup into the HTML file itself.

Is the limit 2 MB or 15 MB?

Both figures are real but apply to different scopes. 2 MB is Googlebot's limit for HTML and text files, which governs Search indexing — the number that matters for SEO. 15 MB is the default file-size limit for Google's broader crawling infrastructure. For your HTML pages, treat 2 MB as the working limit.

How do I know if a page exceeds the limit?

Measure the raw HTML response size (Chrome DevTools Network tab, View Source, or curl | wc -c). Greadme's Deep Scan measures the uncompressed HTML document automatically and flags pages near or over 2 MB.