A french SEO guy claims that Google is able to update its index in 14 minutes. The guy based his comment on a test he made: he published an article on his blog, which was appearing a few minutes later on Google Blog Search, and then directly on Google’s default web results. The guy also backs his comment with what a Google Engineer from Googleplex had supposedly said to a friend of his.
It could have been true if Google hasn’t announced a few months ago the launch of Universal Search, a new feature that consists of displaying different kinds of content such as Images, News, Videos, etc directly into web results.
What the guy noticed was just Universal Search in action, Google has displayed the blog post directly in their web results, and not only in their Blog Search results.
So it was not an update of their entire database, but rather just the displaying of a blog post based on the user’s query.
Source: Seo By the SeaÂ
Hi Nadir,
What interested me about this topic isn’t whether or not Google’s full index is updated that quickly – it really can’t be – it takes to long to crawl and recrawl web pages.
But, what is interesting is people’s perceptions of how fresh the content is at Google. Blog something, and 10 or 15 minutes later, it shows up for a search. That’s pretty amazing, but I’ve experienced many pages(recently) that don’t get updated for a few weeks after changes have been made.
Then again, I really think that blogs were getting indexed this quickly even before Universal Search.
Google blog search, I believe, doesn’t fill indexes by crawling but by RSS feeds (This also explains why feedburner aquisition is important for it) and thus it is no wonder that pages appear fast on blogsearch. Blogs providing partial feeds don’t appear in blogsearch results for keywords that doesn’t appear in feeds but on full page (it’s only when googlebot crawls the page that it appears on results.
I believe that you are correct, Jalaj. Google blog search does appear to be indexing the feeds, and not the pages. And news results appear pretty quickly in Google’s results sets, though once again, they originate in Google’s News repository.
Regardless of whether or not those are crawled or are part of a stream of data acquired through feeds, they are sometimes blended into Web results regardless of the fact that the Web results listed may not have been crawled for weeks or longer. The whole database isn’t getting refreshed every hour or less, but with the fresh looking news and blog pages, it could really look like it is.
Hi guys. Yes, exactly. That’s the power of RSS feeds and pinging 🙂
I believe that Google is able to update its index in 14 minutes. Whole index is stored in MySQL DBs on data center machines with associated connections.
Index of http://www.sooslic.com has 4000000 lines, duration of index update is 5 second.
Hi,
I am the guy you related the story, my name is David Degrelle and I am CEO of 1ère Position SEO-SEA agency and work from 1999 in search marketing industry in France, founder of e-SEMA and vice-president of SEMPO Europe.
From my point of view, Universal search for Google does not have anything to do with what we are talking about : crawl AND ranking in real time by Google.
Moreover I am a little surprised of what you said here and another blogs because identify results coming from the universal search are relatively simple and visible : it is about to read what is written in the organic results of Google 😉
When Google adds organics results from universal research it mentions it, example: “News results for keywordâ€, “Images results for keywordâ€.
However, I never saw appears to date (all can change) “Blogs results for keywordâ€, when Google display results from blogs.
I do not believe, always according to my experiment, that Google actually makes really a difference between a traditional web site or a blog, as long as it judges that the result suggested is most relevant for its visitors, site or blog it does not matter, the stake for Google is to answer more close to the request of a user.
Perhaps more easily, you will agree to the article of Matt Cutts, know as GoogleGuy, that I thank for having established a link towards my original article, and who makes the same reports and without never speaking, him either, of universal research to explain this huge change about Google.
The change that we saw arriving is much deeper than you seem to imagine and far from being as basic as ping google blog search, being in Google news or RSS stuff.
I propose you another article very interesting about Google Bounce Factor on how Google uses many other datas to detect and calculate ranking position always more quickly in its index via the use of the data from Google Toolbar, Google Analytics and Google Adwords, and surely still many other sources more secret than it remains us to be discovered… I have some more ideas about it following how Google fight clic fraud but it’s another story.
14 minutes? If there are more search engine references on the topic of it being indexed in 14 minutes than on the “article” itself, I doubt Google saw fit to expedite the dispatching of the bot to his site and then hastily indexing it’s incredibly important content. I’d say, falling into a time warp would be more likely. That, or a little viral marketing.
Hi David,
Nice “meeting” you. Thanks for commenting.
Matt Cutts did indeed post something about how fast Google was at indexing pages (http://www.mattcutts.com/blog/minty-fresh-indexing/). He didn’t really explained why it happened, nor said how this works: what factors imply such a fast indexing?
Danny Sullivan asked in this same thread if this was related to Blog Search results:
“Rich did a post, on a URL independent of his blog’s home page, which populate out to Google Blog Search and might have shown up as a standalone page within the regular search results. Or not — it’s not clear.”
Matt Cutts replied to Danny and other people who asked if this was related to Blog/news results generated by RSS in a fast way:
“Philipp Lenssen, I honestly haven’t talked to that team to see if RSS/pings are a factor. I just noticed the freshness as a user and thought I’d mention it.
Danny Sullivan, I wouldn’t claim to be an expert on this particular topic, but remember that one of the points of “Big Daddy†was to unify our crawler. Crawling from main web search, blog search, image search, etc. all go through a single crawl interface. My understanding is that the main web index can be minty fresh while being separate from (say) the blog search.”
So he’s “implying” that the main index can be fresh while being independent from blog search, or other vertical search engines.
From what I’ve experienced so far, it seems like only pages that have a RSS feed appear that fast in Google’s main index right after being posted.
This is just a guess. I could be wrong, so, I’m open to hear strong facts that will back this phenomenon.
I’d be happy if you or someone else can point me to a site that behaves similarly, but that doesn’t have a RSS feed.
Yes, it is definitely possible!!
My latest finding about my blog post managed to get indexed within 10 minutes (9 minutes to be exact). You can check out my blog post Google Indexed Post In 10 Minutes for details.
In fact, some blog even mentioned in less than 1 minute. That is the goal I want to achieve for my blog.
Regards,
Steven Wong