homepage Welcome to WebmasterWorld Guest from 184.72.69.79
register, free tools, login, search, pro membership, help, library, announcements, recent posts, open posts,
Become a Pro Member

Home / Forums Index / Google / Google SEO News and Discussion
Forum Library, Charter, Moderators: Robert Charlton & aakk9999 & brotherhood of lan & goodroi

Google SEO News and Discussion Forum

This 64 message thread spans 3 pages: 64 ( [1] 2 3 > >     
Patent Granted To Google's Panda On SERPs Ranking
engine




msg:4657181
 7:29 pm on Mar 25, 2014 (gmt 0)

Hat tip Bill Slawski

Invented by Navneet Panda and Vladimir Ofitserov
Assigned to Google
US Patent 8,682,892
Granted March 25, 2014
Filed: September 28, 2012Patent Granted For Google's Panda On SERPs Ranking [seobythesea.com]

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for ranking search results. One of the methods includes determining, for each of a plurality of groups of resources, a respective count of independent incoming links to resources in the group; determining, for each of the plurality of groups of resources, a respective count of reference queries; determining, for each of the plurality of groups of resources, a respective group-specific modification factor, wherein the group-specific modification factor for each group is based on the count of independent links and the count of reference queries for the group; and associating, with each of the plurality of groups of resources, the respective group-specific modification factor for the group, wherein the respective group-specific modification for the group modifies initial scores generated for resources in the group in response to received search queries. [patft.uspto.gov...]

 

Martin Ice Web




msg:4657208
 8:46 pm on Mar 25, 2014 (gmt 0)

I miss two important words "compelling" and "content"!
So panda is jsut a new way to count links. It is not content related?!

superclown2




msg:4657215
 9:10 pm on Mar 25, 2014 (gmt 0)

Sir Humphrey Appleby would have been proud of saying all that. Any chance of a translation into English?

aristotle




msg:4657216
 9:18 pm on Mar 25, 2014 (gmt 0)

I haven't had time to study it in detail, but one of the factors it appears to take into account is how often people search for the specific name and/or URL of a site.

roshaoar




msg:4657233
 9:42 pm on Mar 25, 2014 (gmt 0)

It sort of reads as if google divides pages or sites into groups and then has a different weighting for each of these groups based on queries and inbound links...

JD_Toims




msg:4657248
 10:03 pm on Mar 25, 2014 (gmt 0)

The patent was granted to Panda, but it looks more like Penguin is the system described -- The date it was filed seems to point to it being Penguin too. Panda was released in Feb. 2011 and it doesn't seem like they'd wait over a year and a half to file the application, but Penguin wasn't released until April 2012. Six or so months the get the application written and filed sounds more reasonable to me -- Anyone know how involved Navneet Panda was with Penguin?

Dymero




msg:4657260
 11:42 pm on Mar 25, 2014 (gmt 0)

I'd note that people seem to be assuming this is the Panda patent because of the name of the inventor, but it could very well be Penguin. I don't think we ever found out who developed that one.

londrum




msg:4657268
 12:04 am on Mar 26, 2014 (gmt 0)

I don't think anyone's going to want to copy panda, it's rubbish

turbocharged




msg:4657389
 11:20 am on Mar 26, 2014 (gmt 0)

One of the methods includes determining, for each of a plurality of groups of resources, a respective count of independent incoming links to resources in the group;

Seems simple enough. Spam the entire group with links to inflate the link count for the group and penalties for websites that have engaged in spam will be revoked. Which websites comprise this group that are most likely being used for the sampling? The websites that are ranking good of course.

We can add this patent to the negative seo concerns that many people have and/or have already been hit with.

roshaoar




msg:4657405
 12:12 pm on Mar 26, 2014 (gmt 0)

Well I think the really interesting thing is this plurality of groups talk, and what it might take (or probably avoid doing) to be in any "AAA" rated group. So much pre-Panda SEO is people looking for shortcuts to fame and glory.

Yet I'm having by far my biggest success with a site I started less than 6 months ago and on which all I'm doing is writing superb unique content, and following Google guidelines, and using intelligent webmastering. To now be in the top 5 for a single word query with 258 million results isn't bad going.

zeus




msg:4657467
 4:03 pm on Mar 26, 2014 (gmt 0)

The dude that created panda, hes name is Panda, I think Matt mentioned that once. I also dont believe in this content thing.

mack




msg:4657470
 4:14 pm on Mar 26, 2014 (gmt 0)

There could be a bit of "smoke and mirror" going on here. This may have been invented, and patented, but we do not know that it was implemented as-is.

Owning a patent prevents others from being able to copy the methed exactly, but it doesn't mean this is how Google are using it. A subtle change can make a big difference to an algorithm.

It gives a little bit of an insight into how things are pieced together, but its not a pot of gold.

Mack.

zeus




msg:4657476
 4:44 pm on Mar 26, 2014 (gmt 0)

I think we get more out of such a Patent note, then anything else, Matts videos included.

flatfile




msg:4657487
 5:08 pm on Mar 26, 2014 (gmt 0)

Now I remember the time before the Panda update, a person from Google said on hacker news something along the lines of "We've changed the way we look at links".

rowtc2




msg:4657499
 5:30 pm on Mar 26, 2014 (gmt 0)

So, Panda is a formula with a number of links and a number of searches for brand (domain) name?

Another interesting fact in patent is "implied links": is not a clickable link, is just text and this can be counted as link.

muzza64




msg:4657515
 6:15 pm on Mar 26, 2014 (gmt 0)

I've suspected for a while that Panda is mostly about link quality and a little bit about other signals that high quality content generates naturally. Maybe this adds some weight to that, I dunno, but it might explain why so much duplicate content on well known sites ranks so well.

So it's kinda about content, but content doesn't matter much if you have a big name behind you.

But there is a new Panda coming which is going to be more friendly to small businesses/sites. Maybe that will be less influenced by link quality and more by content quality (i.e. what we're learning from this patent may already be obselete).

bwnbwn




msg:4657526
 6:38 pm on Mar 26, 2014 (gmt 0)

Interesting read. I know we have all looked up searches related to your nitch. Some you show and some you don't even though from your content/pages you have a page that should. This is all about spreading out the traffic related to a group (nitch). I have websites showing for good terms but won't show for a less trafficked long tail term. It should but it doesn't. Remember Google throttling?
[webmasterworld.com...] internetheaven in 2009
[webmasterworld.com...] Tedester in 2010

This just confirms (to me anyway) what Google tried to do back then. Grouping websites per nitch, sub group, and sub sub groups and rank them based on links in a group, sub group, or sub sub group. This grouping can be extended to allow as many groupings as Google needs to display websites per grouping. This grouping can then be based on the number of search terms each group would receive over a given period of time.

I understand some think this might be or is related to Panda update I don't, and feel it is related to traffic distribution based on your grouping related to competition related to links.

whatson




msg:4657536
 7:29 pm on Mar 26, 2014 (gmt 0)

I think this is Penguin, not Panda. Maybe just to throw people further off the Panda scent. Panda is an algo that you can presumably change by simply making changes to your site. Google would never want to disclose what these "changes" might be, as people can then game their system.

EditorialGuy




msg:4657544
 7:54 pm on Mar 26, 2014 (gmt 0)

Panda is an algo that you can presumably change by simply making changes to your site. Google would never want to disclose what these "changes" might be, as people can then game their system.


On the other hand, Panda (unlike Penguin) supposedly is about quality. Why would Google mind if people improved the quality of their sites for better rankings?

whatson




msg:4657557
 8:16 pm on Mar 26, 2014 (gmt 0)

I said to game their system. e.g. if Google doesnt like ads in this area, then the "gamers" can put them in another area. If google needs a min of 400 words for good content, then they will add 400+ words, etc, etc.

It is a way for Google to identify sites that are genuinely the real deal, as much as possible.

EditorialGuy




msg:4657567
 9:04 pm on Mar 26, 2014 (gmt 0)

If google needs a min of 400 words for good content, then they will add 400+ words, etc, etc.


It's unlikely to be that simple. Still, one can certainly make an argument for not revealing a site's "Panda score" (or even the reasons behind manual penalties).

Andem




msg:4657583
 10:04 pm on Mar 26, 2014 (gmt 0)

I have to concur with whatson. Reading the overview had me thinking about none other than Penguin, not Panda. There will be no doubt assumptions and FUD going around now for months, if not years as a result of this.

whatson




msg:4657615
 12:25 am on Mar 27, 2014 (gmt 0)

@EditorialGuy obviously not that simple but I just wanted to give an obvious example of what I implying.

indyank




msg:4658448
 4:34 pm on Mar 29, 2014 (gmt 0)

I will have to agree that this is Panda and Navneeth Panda is one of the guys who helped google in overcoming the technical challenges that panda implementation posed.

And yes Panda is about links (link quality) and no wondewr not many figured it to recover. The google employee moultano did reveal it on hackernews just before panda was released.

Some past threads to go thro. what people discussed then here on webmasterworls

[webmasterworld.com...]

and read what many of us were discussing in the following thread. in particular read what Tedster had wondered then on this page.

[webmasterworld.com...]

whatson




msg:4658533
 7:55 am on Mar 30, 2014 (gmt 0)

Sorry indyrank, but it doesnt sound like you understand Panda at all.

indyank




msg:4658538
 8:52 am on Mar 30, 2014 (gmt 0)

Ok. if that is your opinion, let it be so. I have seen panda examples right from day 1 (Feb 11 2011), read every word of that patent to understand how it works.I also know why they are not calling it a penalty but a new way of ranking results and why brands tend to do well post panda.

flatfile




msg:4658597
 3:01 pm on Mar 30, 2014 (gmt 0)

@indyank it seems that people somehow forgot that Panda was about links. Probably because Google said "create quality content and you'll recover". The more I think of this whole thing, the more I realize that Panda and Penguin are more related than most of us think.

indyank




msg:4658601
 3:51 pm on Mar 30, 2014 (gmt 0)

On the other hand, Panda (unlike Penguin) supposedly is about quality. Why would Google mind if people improved the quality of their sites for better rankings?


Good question. I had the same question way back then in Feb 2011. But I am now very clear on the reason after reading thro. that patent. It is because the method they use had some loopholes then that would let website owners to game the panda algo. I am sure that they have been plugged now.

whatson




msg:4658636
 7:46 pm on Mar 30, 2014 (gmt 0)

We've all seen Panda examples from day 1. A large part of my portfolio of sites were hit by Panda, they were covered in ads, and rightfully deserved to be hit. But I assure you, the links they had couldn't have been more legit. They were excellent, 100% natural. This is from several sites. Panda was more about user experience. Not one of my sites were hit by Penguin, Penguin is about links, and I had a squeaky clean link portfolio.

Panda = site/content quality
Penguin = link quality

Just think about it, if you were Google, wouldnt you want these two factors in your algorithm.

Planet13




msg:4658672
 9:58 pm on Mar 30, 2014 (gmt 0)

I don't think Panda was limited ENTIRELY to on site factors though.

It seems in the wired interview with Matt and Amit that they tried to quantify ALL the factors that made up "good" sites and those that made up "poor" sites, and used that as the basis of Panda.

Based on that interview and Eric Enge's interview with Vanessa Fox (she said that Panda is a "platform" and not an algo update", Panda was much more than just analyzing content or analyzing links:

"Panda isn't simply an algorithm update. It's a platform for new ways to understand the web and understand user experience."


- Vanessa Fox

Certainly part of that quantifiable data would involve links as well as other off-site factors.

It's also important to note that Panda was made possible by Caffeine infrastructure update, and my understanding is that Caffeine allowed for a significant increase in the ability for google to understand a website's relationship with the rest of the web universe.

This 64 message thread spans 3 pages: 64 ( [1] 2 3 > >
Global Options:
 top home search open messages active posts  
 

Home / Forums Index / Google / Google SEO News and Discussion
rss feed

All trademarks and copyrights held by respective owners. Member comments are owned by the poster.
Home ¦ Free Tools ¦ Terms of Service ¦ Privacy Policy ¦ Report Problem ¦ About ¦ Library ¦ Newsletter
WebmasterWorld is a Developer Shed Community owned by Jim Boykin.
© Webmaster World 1996-2014 all rights reserved