I'd say you're right. Penguin was first released on April 24 - see this discussion [webmasterworld.com]
There has been one and only one Data Refresh for Penguin and no algorithm change. The data refresh was on May 26 - that Penguin Refresh discussion is here [webmasterworld.com] No one one in that discussion reported a recovery.
My take on Penguin is that it is another "sub-algorithm" based on machine-learning, just as Panda was. That means any recovery can only happen when Penguin is re-run or at least has a data refresh.
I've read some reports around the rest of the web that claim some Penguin recovery. However I have never been convinced that Penguin problem was the correct diagnosis in any of those cases. Either the date of the supposed "problem" did not coincide with Penguin or its recovery did not coincide with the one Penguin Refresh.
I don't think a refresh is going to help.
As far as I can see, Penguin was a recalculation of a websites linking pattern with relevance as the base for PageRank calculations.
With the removal of the non-relative link's influence on SERPs, being part of the equation there will not be a recovery.
The other part is a set of penalties based on the anchor text found on the non-relative links.
These are tough penalties.
It seems they will even follow across to a new domain, if the old one is linked, or perhaps if the owner is the same in the whois check.
My guess is that maybe Penguin is all about external links. And since external links take time to build or take down, it doesn't make sense to refresh Penguin every month.
I think there's plenty more going on with Penguin than a major adjustment to the link equity calculations. That alone would not require the complexity or calculation overhead of an independent algo. In fact, there is a lot of evidence that link equity adjustment is going on in an effort that is independent of Penguin.
Also, recent comments from Matt Cutts indicate that the Penguin algorithm has been undergoing major work these past months. Again, such an effort would not be needed if the Penguin algorithm were only about links. Yes, the link profile may be (and probably is) involved. But I'm predicting that when Penguin 2.0 finally hits, we'll see something very confusing and more evidence of Penguin's complexity.
I also see that some sites have recovered lost rankings and traffic by fixing their backlink profile, even though Penguin has not been revised or refreshed.
I think it would.
Just think of all the overhead it would take to recalculate PageRank.
Especially if it were comparing linking to linked pages for relevance and not just using the calculated page rank.
Where do you see this evidence regarding independent link equity outside of Panda.
I gave up believing in Matt Cutts a LONG time ago.
The duplicity of the man is astounding.
|Just think of all the overhead it would take to recalculate PageRank |
As I understand it real PageRank [not toolbar PageRank] is already recalculated continually - daily for most pages on most sites. It's been that way for several years, even before the Caffeine infrastructure went live. Once the math for a change in the PR formula is generated, the job is essentially done and the overhead would be minor, IMO.
I don't want to go far off-topic for this thread, however. We're talking about Penguin recoveries and I really doubt that PageRank is a factor there.
One factor that got some discussion for Panda but almost none for Penguin is the keywords the decide which sites get their Penguin factors calculated. With a machine-learning process, it has to be a subset based chosen in some way.
For example, with Panda version 2.0 was reported by Amit Singhal as going further down the long tale. For Penguin, which is reportedly looking for rankings that are "artificially" boosted, we've got to be talking about using a certain selection of keyowrd rankings to choose where the calculations are made.
|that are "artificially" boosted |
Did anyone publish those artificial items into a definitive list by priority ?
I mean if such a list existed, and remedial steps were taken, why wouldn't sites recover ? [ just throwing out the line to see what comes back ].
Penguin , on the face of it [ hitting over optimized sites ] would seem to be an easier fix than Panda , since there is more subjectivity in Panda, IMO.
That's the issue. Almost[?] no recoveries, so no list exists that we can trust. Not from SEOs and not from Google spokesmen, either. Lots of guesses, some quite well educated - but no dependable list that I know of. Not even a list of generic ideas like the ones Amit Singhal gave us for Panda on how they generated the seed set.
Yes you are right but it is progressive, not retroactive.
When the PR algos are changed the new "rules" come into effect, but they do not overwrite the old.
With the new rules, PR is calculated on the basis of relevance, which was not in the original algos. This was changed in the Mayday update.
There is also no correlation now between PR and SERPs.
With Penguin the linking factors were again turned on, the linking profile for each website calculated retroactively, and then the linking factors turned off again.
You might not thing PR is a factor in Penguin but it explains exactly hat has happened.
It is the keywords that decide the penalty, but it is not based on the percentage of their use, as is supposed by a lot of people.
"Artificially" boosted rankings would be determined by links on non-relevant pages, and the anchor text used on these non-relevant links, would form the basis of the penalties.
|My guess is that maybe Penguin is all about external links. |
My experience is that it is not.
I had four sites that were affected and only one of these just possibly could have been attributed to external linking problems. I think it may have been as much about internal links (over optimisation) than anything else. All I know is that there are no signs of recovery despite several attempts at remedial work.
Let's allow this thread to have some space where people who feel that there ARE Penguin recoveries report on that. Our theoretical discussions about Penguin factors would be better for this thread:
Penguin Recovery Tips - think tank part 2 [webmasterworld.com]
Reminder to those who have forgotten, Penguin is not just all about links. Penguin is about spammy sites [insidesearch.blogspot.com], period.
Another reminder, Google's unnatural linking messages started months before Penguin, so conflating these with Penguin just makes the conversation that much more confusing. The only direct, semi-authorative mention of links being involved with Penguin is Matt Cutt's exchange way too many links [seroundtable.com] comment at SXSW.
Links, and their link to Penguin, was then reinforced by a respected person in the search field who found that sites that were affected by Penguin had poor backlink profiles. Not much rigor went into that analysis because duh, what did he expect? Spammy sites have spammy backlink proiles!
Correlation. Causation. Face. Palm.
And it spread. Like wildfire. Everybody forgot that Penguin was simply about sussing out spammy sites and it became only about links, not also about other factors. We have to look at our sites and practices very, very objectively and can't be in denial while doing so.
I have some pages recovered. I simply got some good quality links to the page that lost traffic and it moved forward. Tried with several pages.
Just stop whining and get more links.
|It's quite possible I've missed a few posts from all the Penguin comments, but I can't seem to recall a post from anyone who claims to have actually made a full, lasting recovery. |
I can report a recovery. One of the sites I work on was hit by Panda, unnatural links, and Penguin (1.0). The site escaped Penguin at 1.1 but it is still affected by Panda and unnatural links.
The only change to the site was a significant removal of unnatural backlinks. This was really done to help the unnatural link penalty, but it seemed to help this particular website escape with the Penguin 1.1 refresh.
subscribing to this thread for interesting comments on BL's and Penguin