Matt Cutts confirms Penguin Data Refresh - Oct 5, 2012

[SnowMan68]

Looks like it finally happened!

Weather report: Penguin data refresh coming today. 0.3% of English queries noticeably affected.

[twitter.com...]
.

[edited by: Robert_Charlton at 1:56 am (utc) on Oct 6, 2012]
[edit reason] fixed link [/edit]

[supercyberbob]

In other words, time to link spam Gooble into the ground.

[aristotle]

Here is a quote from a WebmasterWorld thread [webmasterworld.com ] started by abhishekmishra in August:

Matt Cutts discussed at SES San Francisco to talk on stage and answer questions...
When people asked Cutts about the next Penguin Update he thought: You don’t want the next Penguin update, the engineers have been working hard.
For Penguin:- The updates are going the be jarring and julting for a while.
Webmasters who want to get as much visibility as possible should look at the spectrum of value you’re adding.

From what has been reported so far, this current update hasn't produced much "jarring and julting". So maybe Matt Cutts was talking about something that the engineers are still working hard on, and that we won't see until later.

[SevenCubed]

Oh and the press-release/yellow-pages spam makes my head spin.

The yellow pages .ca version is dominating local search here. Guess what, they are a PPC google adwords advisor. I know someone here who they are raking big money out of. It's not difficult to imagine they are getting preferential treatment in SERPs to catch the potential that slips through google's own hands.

[smithaa02]

Simsi, I don't know that you can create new content to escape Penguin or that google ignores (rather than after the fact penalizes) artificial links.

For our site (roughly), only the home page was hammered...and pretty much for a major money phrase. Other subpages still rank #1 for other terms.

For some period after Penguin came out, you could misspell your money phrase or add some irrelevant modifier like -amazon.com to see how your site ranks without Penguin. They have sense patched this up, but to me this is evidence that Penguin is an after the fact penalty, rather than a nullification of page juice.

Our pages for example ranked quite well before we did any link building for them, so we weren't dependent on the new links to be on page one (as opposed 30 now).

As for creating new content...while Penguin seems to be largely an offsite factor based on bad links (correct me if I'm wrong!), we have heavily modified our site and redone our home page a bunch of times. Nothing has worked.

If anybody has escaped Penguin by adding new pages or modifying content on existing pages, I would love to hear your story. I haven't found a single example of a person that escaped Penguin by changing their content. In fact finding genuine examples of anybody who has escaped penguin at all (and not Panda) is tough to find.

Now if you mean create content on a brand new unpolluted domain...that is a different story. That can work...but is very hard do to do for money phrases (minor phrases are easy to rank for with new sites).

[Wilburforce]

For our site (roughly), only the home page was hammered...and pretty much for a major money phrase.

Mine too.

For my main key phrase, two EMDs have appeared in new positions today (one at #5, the other at #14), and so far that's about it on the top few pages.

[Simsi]

smithaa02 - Now if you mean create content on a brand new unpolluted domain...that is a different story. That can work...but is very hard do to do for money phrases

Agree: it is hard. I wasn't really referring to a new domain to be honest. What I was getting at is that Google simply doesn't like sites being gamed and changing link profiles to react to Penguin does - as someone else has suggested - rather advertise the fact.

I just think that it is in everyone's best interests (long term) to focus on good quality content and if you can't benefit from linking, that is the obvious route to follow. Whether it can be successful in beating a Penguin drop or not who knows, but it makes sense that if Google saw your site get better and users react to that, it figures you could bounce back.

After all, understanding where search wants to go long term is really just applying some common sense and logic, albeit that right now, there are loopholes that we can exploit (I'm deliberately putting aside the Knowledge Engine [webmasterworld.com] stuff here).

----

Moving on, I just checked ranks for a number of sites that I manage and am seeing today's refresh hitting some of them hard. Interestingly, it is hitting the sites which were set up over several years mainly to exploit the loopholes I referenced above. Conversely, a couple of sites that have never really relied on links and get more 'loyal' traffic have risen simultaneously, although I think that is more because sites above them have been penguined.

But that says to me that Google's latest refresh is doing a pretty decent job of spotting sites that are being artificially boosted. Incidentally, the sites that have been hit have survived every Panda & Penguin update until today. They aren't low on quality information but they have been subjected to a lot of SEO down the years (links / EMD stuff mainly).

[seoskunk]

Just suffered a 500 drop at the hands of penguin. I just don't understand this anymore......

[crobb305]

Just suffered a 500 drop at the hands of penguin. I just don't understand this anymore......

seoskunk,
Did the 500 position drop happen yesterday after MC made the announcement, or did it happen today?

I am still trying to determine if the new Penguin refresh has occurred. I recall the first Penguin taking a few days, but it was much more significant than this was supposed to be.

[seoskunk]

Hi crobb305 the drop happened last night, also every sub category bar one has dropped, so pretty bad news all round. Links were I thought safe, just quality directories so I feel like its a draconian update right now.

[smithaa02]

Skunk...out of curiosity what was the anchor text diversity on your directory links?

What was your link velocity? Like say you have 'organic rate' of 3 a week, did it spike up to say 30 a week for brief period of time then fall back down on graphs like ahrefs?

Was the ratio of directory links to non-directory links unnatural?

[crobb305]

I'm seeing some new fluctuations this evening. For example, in my niche, a couple of sites that moved up after MC's announcement yesterday, and gained sitelinks, have dropped back down and lost the sitelinks. For my own sites, I am seeing significant improvement on Google.com.au and Google.co.uk. I am wondering now, if we are going to see an iterative update for a day or two, and if anything we see international could be indicative of things to come. I may be dreaming, but I do see the recent fluctuations on Google.com that I didn't expect (when I assumed Penguin had happened, and settled).

[bluntforce]

Is this just me?

Matt Cutts tweets about a Penguin data refresh but in his tweet there is a link to "Details" which reference Panda on webmaster central.

I don't believe there's any hidden meaning there, but it certainly clouds the issue for me.

[crobb305]

bluntforce, the details are about Penguin (though he did summarize some info about Panda in the 4th paragraph), but the overall post was about their impending webspam update. It was written April 24, and that's the day Penguin 1.0 was released.

[aristotle]

I'm trying to visualize how a data refresh works. Does it mean that all the new data is dumped out at the same time? Or is the data fed out slowly and gradually? And is there a kind of series of iterations that propagate back and forth across the web?

[atlrus]

The home page of one of my UK websites went from page 1 to page 18, starting on the 5th...inner pages still ranking high for now.

[pelizden]

At several of my affected sites, my new content can be only found when put like "my new content", only with quotation marks. Having this in mind, there doesn't seem to be much use in working on affected sites, especially when you consider that some scraping blog might copy your content and rank well for your work.

[Perren]

Has anyone tried to add new links to a penalized site instead of removing old ones?

[blackpixel]

Perren, I tried and no result.

[Wilburforce]

Has anyone tried to add new links to a penalized site instead of removing old ones?

Not manually, but my penalised pages have gained links organically since Penguin 1 (which has had no apparent effect).

I wouldn't expect new backlinks to undo the damage if Google has penalised a page for an unnatural link pattern. My penalised pages have backlinks from a few sites that have each provided - unsolicited - hundreds of links with identical anchor text pointing to the same page, which looks unnatural whichever way you view it.

As I can't disavow them in WMT (unlike Bing) or remove them, Penguin makes me vulnerable to off-site events I can't control, which as well as being inherently unfair has questionable validity in assessing quality or relevance.

Looking at the latest Google SERPS, one might suspect that a Microsoft mole had infiltrated Google's algorithm team.

[Jez123]

Is there no positive movement for anyone? What a waste of 5 months. It would have been more humane of Cutts to just tell us we were already dead.

[Shepherd]

Penguin data refresh reported...

No recoveries reported...

I hope Larry and Sergey leave google INC someday soon and start a search engine.

[Perren]

blackpixel, what kind of links did you add? Different anchor texts? Spread over time?

Wilburforce, strange as that would make negative seo quite easy. Sure it's not panda or something else that's the problem?

[aristotle]

Matt Cutts realized that all of his efforts over the years to stop spammers had mostly failed, and somehow finally concluded that the only way to really stop them is to bring down any site that might have had SEO applied to it or its backlinks. Penguin is the result. But since it is based on a false premise, its main effect has been to lower the overall quality of the Google SERPs.

[smithaa02]

Some idle musings...

We know google is terrified about people reverse engineering their algo. Penguin is one of their prized animals, so what if they scramble or slow down recoveries for that will recover. So google doesn't like to see sites jump from page 30 to page 1 in a day, right? So if they determine your new norm is page 1...they might put a slow releasing lock on your new rankings, just so you don't get instant feedback. Cross my fingers this is true.

Too much of a coincidence that two Pandas, EMD and now a Penguin data refresh come out. By combining algo updates over a short period of time, you make it difficult to reverse engineer. And if you have a really important update coming up...you put out a lot of noise/updates all at once. I think a Penguin algo (not data) update is around the corner.

That so many of us did not recover despite removing bad links (in fact I haven't read a single recovery story) could be an important clue as to how Penguin works. We know they haven't run a algo update...so it's the same algo just with new data.

Yet, if the data is good now...that means that they aren't looking at things 'now'. My hypothesis is that Penguin is all about time. Was your link building velocity over the past X months within standard deviation? In a recursive manner, was the link velocity to the sites that link to you within standard deviation? Did your and your backlinkers get too many backlinks after being hatched? Are backlink profiles (especially to certain keywords) looking like ski hills or are somewhat flat?

If my theory is correct, then your current backlink profile is irrelevant because google has your backlink profile in history and to them that is more important.

[Wilburforce]

Perren - I can't be certain it was Penguin, but the timing of page shifts coincided with Penguin 1 & 2, and yes, penalising on the basis of backlinks makes negative SEO easier (there have been a number of posts in earlier threads on this).

Yet, if the data is good now...that means that they aren't looking at things 'now'.

Not necessarily. Your "slow-releasing lock" would be one mechanism that could account for it. Whatever the case Google have successfully obfuscated the mechanism (and, in my view, any logic that might previously have been apparent in the results).

[aristotle]

If my theory is correct, then your current backlink profile is irrelevant because google has your backlink profile in history and to them that is more important

.

If you manage to get a lot of old backlinks removed, especially backlinks from a number of different domains, that could indicate to Google that they were under your control, and strengthens the evidence that you did some artificial link-building in the past.

[Sgt_Kickaxe]

Is this just me?

Matt Cutts tweets about a Penguin data refresh but in his tweet there is a link to "Details" which reference Panda on webmaster central.

I don't believe there's any hidden meaning there, but it certainly clouds the issue for me.

Smoke and mirrors tend to do that. Imo just focus on your link graph, produce high quality linkable content and stay the course. Things like Matt Cutts using twitter instead of G+ are just.... distractions.

[Leosghost]

Things like Matt Cutts using twitter instead of G+ are just

Matt making sure that there was actually someone there on the "platform" to notice .. :)

[seoskunk]

Matt making sure that there was actually someone there on the "platform" to notice .. :)

LOL that cheered me up from my 500 plus demotion. 0.3% of the net I guess I should be considered elite at something :)

[spunkle]

2 of my sites recovered from penguin with this update.

Hopefully the data refreshes will be more frequent from now on.