Google Operating System points out something interesting in Google’s algorithm recently: a preference in favoring recent content.
The example provided would seem to favor the conclusion; TCP/IP’s anniversary today has resulted in Google preferencing recent posts, including from Digg, over informative articles related to the search term such as Wikipedia who would have normally had the top or near to the top position.
Indexing recent posts has been a strength for Google, to the point that at least for areas like Blog Search they’ve become the defacto standard as others such as Technorati have struggled to keep up. That Google would preference recent posts in its search results without the usual incoming authority links throws out the rules we’ve always known with Google, and I’m not sure if this is a good thing or bad thing. It would depend a lot on the results, but it would limit attempts to game Google results through incoming links. Perhaps it’s the first step towards Google embracing Web 3.0 with semantic search that learns as it goes, constantly updating its results to suit the user at the time they are searching, complete with contextual awareness as well.
This is very old news in the SEO world.
Happy New Year, America.
Mark
we’ve been seeing this over time, but GOS makes a great point that it’s stronger now than ever.
GOOD! Anything Google does to abate its epidemic of Wikipedea SPAM is a godsend.
Each time this happens, I end up being a winner… hopefully this trend continues yet I am sure there will be a lot of bitching on the boards across the internet that they can’t be found anymore (ei: ranking went down).
Jon
Duncan,
True. If you were really smart you’d have titled this post January 1 tcp/ip. Perhaps f I use January 1 tcp/ip enough times in this comment then perhaps you’ll show up as for January 1 tcp/ip as number 1 thanks to the logo search January 1 tcp/ip that Google has to celebrate the fact that tcp/ip has a birthday of January 1.
Did I mention that I love January 1 tcp/ip?
Mark.
* * http://blogosco...rum/119648.html * *
This post may have been one of the sparks for this debate and re evaluation.
But there is another answer that is equally likely -
Google may in fact be calculating backlinks in real-time – but not displaying them in the organic SERPs by using the LINK: operator.
However, if you use the link operator in the BLOGSEARCH site, you will see almost real-time links that you can further analyze by dividing the time periods to filter them out by hour, days or weeks.
This is probably integrated into their organic SERPs in a matter of minutes.
A few years ago, it took about a day for this effect to happen – and a few years before THAT, there was only monthly updates.
Just imagine how much their algo-hi-tech will improve by the end of this decade
Will this actually help Technorati? I wonder…
I was wondering why they added the “tcp/ip” in their logo search link.
Argh. I think I am still stuck in web 1.0 and trying to catch up. I think I will start anticipating web 4.0 instead…. Maybe skip a few generations to catch up,
Does anyone has any news about crawling silverlight from Google search?
As several have pointed out, google started emphasizing recent posts a while ago – 6 months maybe? But typically it’s just a few posts at the top… and they have to be very recent. In other words, any ‘preference’ given to those results is VERY TEMPORARY. It’s not as though the ‘newest’ results stay high in the SERPs, they expire. So as long as spam is controlled, what’s the problem?
Is Web 3.0 real? Want to know more and a quick Google search wasn’t fruitful. Can someone kick down a link to a creditable article?
Pease god, do not try to start talking up web 3.0. Please?
web 9.0 search the web thru a artificially created cortical implant propagated to your brain ingested initially as nanoprobe drops to your ears =) thereby unleashing the power of the web in your brain.. includes embedded subliminal spam filters… avoid spams in your dreams!
LOL
People say this is old news. At the same time, some SEOs are still getting Wikipedia spam results: http://www.wolf...wikipedias-ass/
As poster 12 pointed out, the new posts in the SERPs must not be lasting long at all. I’d love to see something fresh, anything, besides another wikipedia page.
I wonder since when it changed. I have been seeing fresh content given a preference for some time now. It doesn’t stay in SERP for long though. SEOmoz had a great post on this. I will try to find it
As a website owner I will be watching the effect of this change with interest. I have no complaints if I get more traffic.
Yes I have noticed that with my startup, where the most linked site’s post was Dec. 12th. But, posts thereafter(Dec. 18) appear 1st before the most linked.
(Thanks every1 for the coverage).
Spiders are not an effective means by which to search websites in real time (or something close to contemporaneity).
Google’s results are often irrelevant for social and dynamic websites. It’s best for static websites.
This is old news.
Since “Google Universal Search” was launched on May 16, 2007, Google crawls (and indexes) websites with fresh contents in order to show them within the SERPs.
Sometimes you can see within the results 5-minute-old pages.
Of course, these contents are best ranked because they match better the Google relevance parameters (keywords at title, keywords at URL, …) and published on websites with relevance (if not, they’re not crawled so fast).
I thought it was Google just lowering the overall value of PR – but how come some blogs are down and some are not?May be networking is the issue here – I think Google may have decided to this as many networks interlink their sites when they are not related, they are going to penalize them (as if they were a paid-for directory).
I’d love to hear more on this by Google and what they are doing or intend to do by this, rather than leaving everyone to guess.
Parul
http://www.bhopu.com
The current problem with Google’s instant indexing is sourcing authority sites. In the TCP/IP example a savvy spammer setup brand new websites (blogs) which should have zero authority, and zero backlinks which would normally exclude from the index.
Websites with no quality incoming links should not be ranked #1, as this example painfully illustrates.
Google engineers will hopefully fix this soon so we won’t have to worry about zealous wikipedia hackers taking over more search results positions.
The example provided is so specific that perhaps there just wasn’t that much old content to reference? I did a search for “SES New York”, because in the past it has proven very hard to get Google to come up with *upcoming* events rather than past events. That is still true; although Yahoo and Ask offered SESNY 2008 as the first result, Google shows it nowhere. Since assumedly pages that describe the 2008 event are more recent than those describing 2007, that seems to run counter to the idea that recent posts are favored…right? Maybe it’s more along the lines of treating blog posts like news, and has some previous posters have noted, showing the results only briefly?
If done correctly this can be very good when searching for events and other topics were recent is better. If the boost is temporary, to give an opportunity to new sites, the only problem I see are spammers creating new sites all the time.
If done correctly this can be very good when searching for events and other topics where recent is better. If the boost is temporary, to give an opportunity to new sites, the only problem I see are spammers creating new sites all the time.
Yes this may be old as far as them placing such a quick crawl on new posts on a blog, or a news site, however this is something that does warrant extra attention.
Add in the other big changes over the last 2 quarters at the plex and you will see a bigger picture. There was a PR re-assignment in recent months, paid links became the target of intense prejudice with the google webspam team, they experimented with a digg like feature for the SERPs annonced on the Google Labs blog. Those 4 things tell me that we might be getting another big fat algo change by spring. Why?
Google is nefarious for testing things in an incremental fashion, piece by piece. Each of these things are separate pieces that are being fine-tuned. Put it all together, and voila.. an algo change.
This has been known for some time. Everything evolves though!
I’ve noticed this phenomenon a few a our sites. New pages and recently indexed pages seem to show up quickly in the SERPs, then dwindle over the next few days. It’s a cool feature if it’s true, because it allows smaller, less established sites to get noticed.
what the frick does ‘newness of post’ have to do ’semantic search’?
I was wondering why they added the “tcp/ip” in their logo search link.
This is interesting. How can an SEO use this information to ones advantage? To me it looks like blog sites now have a distinct advantage in terms of SEO. This scares me, someone badmouthing our company in a blog is likely to outrank our website in the serps when searching for our company name (maybe for a short period).
It is time google was made to be accountable. I dont understand how anyone let google become so powerful. In an instance they can ban your site from the google index without telling you why … and there isnt a damned thing you can do about it. This sort of thing can and has closed companies.
I would like to see information about website bans being made public – via osme form of forum. This way the site owner and the community in general would be able to comment and / or resolve issues.
I wonder if this change is somehow related to the really weird subcategories that show up under the main Wikipedia link if you type “wikipedia” into Google: http://www.more...-subcategories/
now we are in end of Web 2.0 …
But what do we have in our hands? Nothing…
and u saying Google be in Web3.0…. I suppose its wrong. Google is not ready for Web 3.0 , too
am I right?
Good.
Google is try to touch the sky limits…………
We have also seen the changes for some of our top competitive keywords.
While it is true that Google has started giving value to recent content, the quality of the content is very important. Syndicating content and making it resource rich is the key to better search engine rankings
Here you REALY found some SEO NEWS !
Hier findest Du echte SEO News !
Suchmaschinen Faq / SEO FAQ
http://www.ecom...optimierung.php
Google Algorythmus News
http://www.ecom...-die-luegen.php
Lots of talk and discussion, but not much fundemental change in how Google ranks websites
And what does it mean for my online-activities?
not relay posible
how can you prove your seo news?
Always remember for Google: Content is king !
Google says: Content if everything is what you need.
I have noticed Google changing results latley. IU know they change often but the results are messed up. If you type in Roofing St Louis I will have a different result hourly. I am not sure what going on but anyone with advice??