down graph

Helping webmasters diagnose website traffic issues.

hire an SEO professional

Stop speculating about Penguin! Here’s what Google says.

Google Penguin SpeculatorWe have been waiting for a long time now for a Penguin refresh to happen.  Penguin initially rolled out on April 24, 2012 and had a minor refresh on May 25, 2012.  Now, here we are 4 months later and we haven’t had another Penguin refresh yet – that we know of!

Yet, there are so many posts from webmasters telling people how to recover from Penguin! In my opinion, this is dangerous because no one really knows what is necessary for Penguin recovery. There is a lot of speculation circulating amongst SEOs these days.

I have heard a number of webmasters speculating about some of the following things:

1. Is Penguin really about backlink quality?

2. Can you recover from Penguin?

3. Is there ever going to be a Penguin refresh?

4. Do you need to have a Penguin refresh in order to recover?

Rather than add to the speculation, I thought I would research and find out exactly what Google says (primarily via Matt Cutts, head of webspam) about these issues.

Is Penguin really all about links?

If you’ve been affected by the Penguin algorithm, it is because of a poor backlink profile.  This was confirmed by Matt Cutts via Twitter.  A Twitter user, Josh Bachynski tweeted the following:

Twitter Question to Matt Cutts about Penguin

And here is Matt’s reply:

Matt Cutts tweet about Penguin

So Matt confirmed that links are the primary thing to keep an eye on in regards to Penguin.  He did say, “primary thing” though which implies that there can be other factors.

If this is the case, how can a site recover, or perhaps a better question is – can a site recover from a Penguin hit?

Is it possible for a site to recover from Penguin? If so, how?

I have seen a lot of speculation on this in the SEO forums.  Many people are advocating removing links, getting anchor text changed from keywords to urls or brand names, or even diluting the anchor text by creating new links using more natural anchor text such as  “click here” or “visit website”.

The problem is that we have really not had many validated cases of Penguin recovery.  So far the only case that I am aware of is WPMU which recovered on the May 25 refresh after removing a large number of sitewide footer links.  One case is not enough to give us good data to be advising people on how to recover.  So, rather than speculating, what does Google have to say about the matter?

Here is a quote from Matt Cutts at SMX advanced when asked the following question by Danny Sullivan:

Danny: If you were hit by Panda and Penguin, should you just give up?

Matt:  Sometimes. Sometimes you should. It might be better to start over. But both are algorithmic, so if you change the signals, you can recover. If you produce content people like, you can come back. We’ve seen sites come back, like WPMU. But if you’ve been spamming, don’t expect that to work anymore.

This tells us that, yes, recovery is possible.  So what does it mean to “change the signals”? Here is some information that was published on the Google Webmaster Central Blog on the same day that Penguin was unleashed:

In the pursuit of higher rankings or traffic, a few sites use techniques that don’t benefit users, where the intent is to look for shortcuts or loopholes that would rank pages higher than they deserve to be ranked.

We see all sorts of webspam techniques every day, from keyword stuffing to link schemes. that attempt to propel sites higher in rankings.

Let’s stop here and look at link schemes.  Here are some examples of link schemes that Google has supplied:

  • Links intended to manipulate PageRank.  – This is a very wide definition, but I interpret it to mean that if you were personally able to make a link and did so with the intention of improving your site’s SEO then this is a link scheme.
  • Links to web spammers or bad neighborhoods on the web.
  • Excessive reciprocal links or excessive link exchanging (“Link to me and I’ll link to you.”) – The problem here is that there is no definition of the word “excessive”.  Some reciprocal linking is still ok.
  • Buying or selling links that pass pagerank.

Here are some things that we don’t know:

  • We think that having a high percentage of keyword anchor texted links (as opposed to url and brand name links) can trigger the Penguin filter, but no one knows whether this is true and if so, what that percentage is.
  • Many think that diluting your backlinks by creating new links containing anchor text using your url, brand name or generic text like “click here” will help.  There is no official evidence for this.  My gut instinct is that creating more links using the same techniques as you did to create the spammy links is not a good idea.

Danny Sullivan also mentioned the following quote from Matt Cutts in regards to Penguin:

If you’ve cleaned and still don’t recover, ultimately, you might need to start all over with a fresh site.

This means that there are some sites that are not going to be able to recover from Penguin, even after a refresh happens.

Is a Penguin refresh ever going to happen?

Because it has been several months since the last known Penguin refresh, some people are speculating that either a refresh will not happen or that they are happening and Google is just not announcing them.

So, what does Google have to say?

Here is another tweet from Matt Cutts on the day that Penguin initially refreshed on May 25, 2012:

Matt Cutts tweets about Penguin

Matt stated that this was the first Penguin refresh which implies there will be more.

Also, Barry Schwartz posted about a conversation with Matt Cutts here.  In it, Matt speaks about the upcoming Penguin refresh:

…people were asking me when the next Penguin update would happen, as if they expected Penguin updates to happen on a monthly basis and as if Penguin would only involve data refreshes.

…we’re still in the early stages of Penguin where the engineers are incorporating new signals and iterating to improve the algorithm. Because of that, expect that the next few Penguin updates will take longer, incorporate additional signals, and as a result will have more noticeable impact. It’s not the case that people should just expect data refreshes for Penguin quite yet.

In other words…Penguin refreshes are coming!  Be patient!

Does there need to be a Penguin refresh in order for a site to recover?

I’ve heard a number of people speculating that you can recover from Penguin even if there isn’t a refresh.  I’ve also seen a number of case reports of so called Penguin recovery.  However, with the exception of WPMU, most, if not all of these so called recoveries seem to be due to other factors.  Many of these sites likely had partial recoveries from Panda issues rather than Penguin factors.  (If you are aware of a good Penguin recovery story, please let me know in the comments.)

Here is a video where Matt Cutts speaks about manual penalties and algorithms (such as Penguin):

I’ll quote some of what he is saying:

If your site is affected by an algorithm, for the most part, if you change your site, whatever the characteristics are that’s flagging or triggering or causing us to think that you might have – keyword stuffing, or whatever, if you change your site, then after we’ve recrawled and reindexed the page and some period after that when we reprocess that in our algorithms it should be able to pop back up, or, you know, increase in its rankings.

I would think that when Matt is saying, “reprocess that in our algorithms” is the same thing as saying when the Penguin algorithm refreshes (assuming it was Penguin that affected your site.)

Also, in Danny Sullivan’s interview with Matt Cutts, Danny says the following:

If you do clean things up, how will you know? Ideally, you’ll see your traffic from Google recover, the next time Penguin is updated.

That leads to another important point. Penguin, like Panda, is a filter that gets refreshed from time-to-time. Penguin is not constantly running but rather is used to tag things as spam above-and-beyond Google’s regular spam filtering on a periodic basis.

What I couldn’t determine, however, is whether Matt himself was saying that Penguin needs to be updated in order for a site to recover, or if this was Danny’s interpretation.

Regardless, when you put this information together, it seems clear to me that a Penguin refresh is necessary in order for a site to recover.


Hopefully we will get a Penguin refresh soon. When that happens we can sort through the hordes of “I recovered from Penguin!” blog posts and see if we can put some good data together. Until then, be very careful when advising webmasters how to recover from Penguin!

Your thoughts?

What do you think about these Penguin issues? I’d love to have your comments below:

This entry was posted in Penguin Information. Bookmark the permalink.

12 Responses to "Stop speculating about Penguin! Here’s what Google says."

Leave a reply