How to Check if Google Indexed Your Content: Passage Indexing, Crawl Limits, and What Actually Matters for SEO
Most SEOs are asking the wrong question about Google’s crawl limits. They’re obsessing over whether Googlebot processes 2 megabytes or 15 megabytes of HTML — when the real question is far simpler: Is your content actually showing up in search results?
A recent exchange between Google’s John Mueller and the SEO community on Bluesky cut through the noise with a refreshingly practical answer. But the implications go deeper than a quick indexing check. Here’s what the conversation really means for your content strategy — and what most people are getting wrong.
The 2MB vs. 15MB Crawl Limit Confusion, Explained
The debate erupted after Google updated its crawler documentation in early February 2026. Search Engine Journal reports that Google moved default file size limits from the Googlebot page to its broader crawler documentation, since they apply to all of Google’s crawlers and fetchers. The Googlebot page now lists a 2 MB limit for HTML and supported text-based files, and 64 MB for PDFs. The crawling infrastructure docs list the default for all crawlers as 15 MB.
This created understandable confusion. Which number matters? The answer depends on context, and the distinction is more nuanced than most commentators acknowledge.
The 15MB limit still exists. It applies to the general crawling infrastructure — that covers everything: Search, News, and Shopping. But the 2MB limit is specific to the Google Search crawler processing HTML. As Breakline Agency explains, they are just being more precise about which limit applies to which specific process. It is a documentation cleanup rather than a sudden destruction of capacity.
Here’s the critical detail that Google’s official Googlebot documentation spells out: when crawling for Google Search, Googlebot crawls the first 2MB of a supported file type, and the first 64MB of a PDF file. From a rendering perspective, each resource referenced in the HTML (such as CSS and JavaScript) is fetched separately, and each resource fetch is bound by the same file size limit that applies to other files. Once the cutoff limit is reached, Googlebot stops the fetch and only sends the already downloaded part of the file for indexing consideration.
One point that many analyses overlook: the file size limit is applied on the uncompressed data. If you’re checking your page size in browser DevTools, make sure you’re looking at the uncompressed transfer size, not the gzipped version your server sends over the network.
Why the Crawl Limit Panic Is Overblown
Here’s what most people get wrong about this: they hear “2MB limit” and immediately worry their long-form content is at risk. But they’re confusing total page weight (which includes images, scripts, fonts, and video) with raw HTML size. These are fundamentally different measurements.
According to the HTTP Archive’s 2024 Web Almanac, the median mobile homepage loads 18 KB of HTML, 73 KB of CSS, 111 KB of fonts, 558 KB of JavaScript and 900 KB of Images. That means the median page’s HTML is roughly 1/110th of the 2MB limit.
Search Engine Roundtable’s coverage of Mueller’s response included data making this even clearer: the median on mobile is at 33KB, and the 90th percentile is at 151KB. This means 90% of the pages out there have less than 151KB of HTML.
To put this in perspective, Breakline Agency noted that if you take a standard novel — say something around 400 pages — that is roughly what 2MB of text looks like. It is about 2 million characters. Unless your single web page contains more text than an entire novel, you’re fine.
That said, a blanket dismissal isn’t entirely fair either. As Hiland SEO points out, certain edge cases can push pages toward the limits. Single-page applications with inline everything — if you’re embedding massive amounts of JavaScript or CSS directly in your HTML rather than external files — could approach the limits. Data-heavy pages with inline data — product pages with thousands of variants, location pages with embedded JSON, or documentation sites with entire codebases in the page — can bloat HTML significantly. Pages using Base64-encoded images via data URLs are also included in the HTML file size.
Think of it like a carry-on bag limit on an airplane. For 99% of travelers, it’s generous. But if you’re the person who tries to stuff a sleeping bag, a guitar, and three winter coats into one bag, you’ll run into problems — and the solution isn’t a bigger overhead bin. It’s better packing strategy.
Mueller’s Simple Method to Check Passage Indexing
This is where Mueller’s response becomes genuinely actionable. Rather than obsessing over byte counts, he offered an elegantly simple test. As Search Engine Journal reported, Mueller stated: “The way I usually check is to search for an important quote further down on a page – usually no need to weigh bytes.”
That’s it. Take a distinctive phrase from the lower portion of your page — ideally something unique enough that it won’t appear on other sites — wrap it in quotation marks, and search for it on Google. If it appears in the results, Google has indexed that passage. If it doesn’t, you may have an issue worth investigating.
This technique is powerful because it tests the outcome directly, rather than trying to reverse-engineer the process. It’s the difference between measuring ingredients by weight versus simply tasting the food.
How to Run This Check Effectively (Step-by-Step)
While Mueller’s tip is straightforward, there’s a right and a wrong way to do it. Here’s the method I recommend for a thorough verification:
Step 1: Pick 3–5 unique phrases from different sections of your page — beginning, middle, and end. Use phrases at least 6–8 words long that are specific to your content.
Step 2: Search each phrase in Google using exact-match quotation marks (e.g., "your unique phrase from the bottom of the page").
Step 3: If your page appears for phrases near the top but not the bottom, that may indicate truncation. However, before panicking, consider that Google simply may not have chosen to surface that particular snippet — absence from a quoted search result doesn’t always mean absence from the index.
Step 4: Cross-reference with Google Search Console’s URL Inspection tool, which shows whether a URL is indexed and what Google sees in the rendered HTML.
Step 5: For a technical audit, use tools like Screaming Frog to check your raw HTML file size across the site, filtering for any pages approaching 1MB or more of uncompressed HTML.
Passage Ranking vs. Passage Indexing: A Crucial Distinction
The original discussion references “passage indexing,” but this terminology has been a source of confusion since Google first announced the feature. Understanding the difference matters.
As Ranked.ai explains, ranking passages is not an indexing change, despite the wording used to describe it. This distinction matters because indexing and ranking refer to different things. Indexing is the process of storing, retrieving, and crawling information. The confusion stems from the term Google originally used to announce this feature. Since then, they have clarified several times that they rank passages but do not index them.
Google’s own ranking systems guide explains that “Passage ranking is an AI system we use to identify individual sections or ‘passages’ of a web page to better understand how relevant a page is to a search.”
In other words, Google doesn’t index individual passages as standalone entities. Passage-based ranking measures excerpts when responding to a query. It does not mean Google stores websites in discrete chunks.
Search Engine Roundtable reported Mueller’s earlier explanation that passage ranking is not about ranking a specific passage but understanding the content on a really long, not SEO-optimized page, and ranking that page (not the passage) for a query where the passage is relevant.
This distinction is more than semantic. If you think Google indexes passages independently, you might conclude that every section of a long page competes separately — leading to flawed content architecture decisions. The reality is that passage ranking enhances the page’s ability to surface for more specific queries, which is actually an argument in favor of comprehensive content when done well.
What This Means for Content Strategy
The crawl limit debate is really a proxy war for a bigger strategic question: should you write long, comprehensive content or break topics into shorter, focused pages?
The honest answer? It depends — and anyone giving you an absolute rule is oversimplifying.
Long-form, comprehensive content has a genuine advantage under passage ranking. As Neil Patel explains, with passage ranking, Google can pull out sections of your content and rank them independently of the rest of the page. For example, your section on “affiliate marketing tools” might rank for that keyword, even if the entire post isn’t optimized for it. A single well-structured 3,000-word guide could rank for dozens of long-tail queries, each corresponding to a different passage.
But this only works if the content is genuinely well-organized with clear headings, logical structure, and passages that can stand on their own as coherent answers. A sprawling 5,000-word post with no subheadings and stream-of-consciousness paragraphs won’t benefit from passage ranking — it will just confuse both users and algorithms.
Here’s my framework for deciding between comprehensive and split content:
Go comprehensive when: your audience benefits from seeing how subtopics connect, the topic is complex enough to warrant a deep dive, and you can structure the content with clear H2/H3 sections that each answer a specific question.
Split into separate pages when: each subtopic has distinct search intent (informational vs. transactional, for example), the combined length would sacrifice readability, or different subtopics target fundamentally different audiences.
The key metric isn’t page length or byte count — it’s whether a user who lands on any section of your page gets what they need without feeling lost or overwhelmed.
The Modern Indexing Landscape: Beyond Crawl Limits
There’s a broader context here that’s worth considering. In 2026, the bigger indexing challenge isn’t file size — it’s whether Google decides your content is worth indexing at all. As ClickRank reports, Google indexing errors happen when Google discovers or crawls a page but decides not to store it in the index. This topic matters more than ever in 2026 because Google no longer tries to index everything — it filters aggressively based on value, trust, and usefulness.
Mueller has explained that Google crawls pages to evaluate them, not to promise inclusion. After crawling, Google decides whether a page adds enough value to deserve index space. With tighter resource controls and AI-driven quality checks, indexing is selective by design.
This means the more productive use of your time isn’t worrying about whether Googlebot can technically access the bottom of your page. It’s ensuring the content there is unique, valuable, and clearly worth indexing in the first place.
With the expansion of AI Overviews in search results, passage understanding is arguably more important than ever — not for traditional blue-link rankings, but because Google’s AI systems are actively pulling specific passages to construct answers. Well-structured, clearly written passages aren’t just good for passage ranking; they’re the raw material Google uses for AI-generated summaries.
Practical Audit Checklist for Passage Indexing
If you want to be thorough about indexing health, here’s what actually deserves your attention. First, verify passage indexing using Mueller’s quoted-phrase search method across multiple sections of your important pages. Second, check raw HTML file size (uncompressed) using your browser’s DevTools Network tab filtered by “Doc” — if any page is approaching 1.5MB of pure HTML, investigate why. Third, move inline JavaScript, CSS, and Base64-encoded images to external files where possible to keep HTML lean. Fourth, ensure critical content and structured data appear early in your HTML source, rather than being buried under thousands of lines of script tags. Fifth, use Google Search Console’s URL Inspection tool as a secondary verification that pages are indexed as expected.
FAQ
How do I check if Google has indexed a specific passage on my page?
Search for a unique, exact-match phrase from that section of your page using quotation marks in Google (e.g., "your unique phrase here"). If the page appears in results, the passage has been indexed.
What is Googlebot’s actual crawl limit for HTML in 2026?
Googlebot crawls the first 2MB of HTML files when indexing for Google Search. Google’s broader crawling infrastructure has a 15MB default limit, and PDFs get a 64MB allowance. These limits apply to uncompressed data.
Is 2MB of HTML enough for most websites?
Yes. According to the HTTP Archive’s Web Almanac, the median HTML document size across the web is approximately 18–33KB. Even at the 90th percentile, HTML page sizes are typically under 151KB — far below the 2MB limit. Only pages with extreme inline data, scripts, or encoded assets would approach this threshold.
What’s the difference between passage ranking and passage indexing?
Passage ranking is the correct term. Google does not index individual passages separately. Instead, it uses AI to identify relevant sections within a fully indexed page and considers those passages as a ranking signal when matching specific queries.
Should I split long-form content into separate pages to avoid crawl limits?
Not for crawl limit reasons — file size almost certainly isn’t an issue. Split content when different sections serve different search intents or audiences. Keep it comprehensive when the connected structure adds value for the reader.
Does HTML size affect page speed and rankings?
HTML size alone rarely impacts performance, but pages bloated with inline scripts or encoded assets can slow rendering. Page speed is a ranking factor, so if excessive HTML size degrades load times, it could indirectly affect rankings.
Obsessing over crawl byte limits is the SEO equivalent of worrying whether your library card has a checkout limit while ignoring whether anyone actually wants to read the books you’re writing. Mueller’s advice reframes the conversation where it belongs: test whether your content is indexed, ensure it’s genuinely useful, and let the byte-counting anxiety go. The pages that struggle in search in 2026 aren’t failing because they’re too big — they’re failing because they aren’t valuable enough to make the cut.
