Google Admits the Web is ‘Theoretically Impossible to Crawl’

For all its technological sophistication and prowess, Google is fundamentally flawed. The sheer size and complexity of the web has made it fundamentally impossible for any search engine to crawl it in its entirety.

In response to a question posted by an SEO on Reddit, Google’s John Mueller admitted that the way his company crawls the web is neither efficient nor objective. He also said that crawling the web is a difficult and expensive task, particularly when attempting to contend with the vast volumes of spam doing the rounds.

Specifically, the SEO submitted the following question on Reddit:

“Why SEO tools don’t show all backlinks?”

In response, Mr Mueller went into great detail about the limitations of the major search engines, and how even the most sophisticated crawlers cannot possibly reach every corner of the web.

Here’s the full response hosted by Mr Mueller on Reddit:

“There’s no objective way to crawl the web properly.”

“It’s theoretically impossible to crawl it all, since the number of actual URLs is effectively infinite. Since nobody can afford to keep an infinite number of URLs in a database, all web crawlers make assumptions, simplifications, and guesses about what is realistically worth crawling.”

“And even then, for practical purposes, you can’t crawl all of that all the time, the internet doesn’t have enough connectivity & bandwidth for that, and it costs a lot of money if you want to access a lot of pages regularly (for the crawler, and for the site’s owner).”

“Past that, some pages change quickly, others haven’t changed for 10 years — so crawlers try to save effort by focusing more on the pages that they expect to change, rather than those that they expect not to change.”

“And then, we touch on the part where crawlers try to figure out which pages are actually useful. The web is filled with junk that nobody cares about, pages that have been spammed into uselessness. These pages may still regularly change, they may have reasonable URLs, but they’re just destined for the landfill, and any search engine that cares about their users will ignore them. Sometimes it’s not just obvious junk either. More & more, sites are technically ok, but just don’t reach “the bar” from a quality point of view to merit being crawled more.”

“Therefore, all crawlers (including SEO tools) work on a very simplified set of URLs, they have to work out how often to crawl, which URLs to crawl more often, and which parts of the web to ignore. There are no fixed rules for any of this, so every tool will have to make their own decisions along the way. That’s why search engines have different content indexed, why SEO tools list different links, why any metrics built on top of these are so different.”

Food for thought and confirmation that even Google might not be quite as powerful or omnipotent as most assume it to be.