Penguin: Core, realtime and updated today

[Shaddows]

[webmasters.googleblog.com...]

Penguin is now real-time. Historically, the list of sites affected by Penguin was periodically refreshed at the same time. Once a webmaster considerably improved their site and its presence on the internet, many of Google's algorithms would take that into consideration very fast, but others, like Penguin, needed to be refreshed. With this change, Penguin's data is refreshed in real time, so changes will be visible much faster, typically taking effect shortly after we recrawl and reindex a page. It also means we're not going to comment on future refreshes.

[NickMNS]

Awesome, this means that we can start buying/selling links again.

If the links are spammy they wont count but they wont hurt. So the strategy is by 100 links, from a wide variety of sources. 50% may turn out to be useless but you will still get a boost from the remaining 50%. You'll get less bang for your buck, but still more than not buying any at all.

[glakes]

NickMNS, I see things different. Instead of Google ranking those who have the least penalties, their focus can return to ranking content and with a lower reliance on easily manipulated signals such as links. We as webmasters can focus on our content and doing what is best for our customers. I see this as a good thing, and I'm one of Google's harshest critics out there.

[NickMNS]

@glakes

I'm one of Google's harshest critics out there.

Really I would never have thought that of you :o)

I agree with you, my post was made with a certain degree of sarcasm. I seriously doubt that what I described would in fact be possible as I can't imagine that Google would not have thought of such a scenario.

I would assume that there must be some mechanism in place to catch such behavior. It may be as simple as the quality to spam ratio of purchased links being very low, so low that you would have to buy a very large number of links. From those links say 1000 only 5% would provide value, and 950 would be spam. The large volume of spam would then trigger a manual penalty.

It will be very interesting to see how this plays out.

[Simon_H]

Gary has also said that P3 will be removed from all sites over the next few days.

This all seems too good to be true, so it'll be interesting to see how it actually plays out. Given the removal of the punitive side and adding of the granular side, what we may see are sites previously hit by Penguin recover to some extent, but also sites previously not hit by Penguin pulled down to some extent because chunks of their links will be devalued where in the past they hadn't hit the Penguin threshold.

Also interesting to see if this is a new strategy for Google, and they will also drop the punitive side from Panda. Certainly, if they can't justify punishing sites for link spam any more, it's even more unnecessary to punish sites for Panda-type issues. Just filter out the offending pages and leave it at that.

[aristotle]

Also interesting to see if this is a new strategy for Google, and they will also drop the punitive side from Panda. Certainly, if they can't justify punishing sites for link spam any more, it's even more unnecessary to punish sites for Panda-type issues. Just filter out the offending pages and leave it at that.

That's not true. There's a lot more justification for Panda than for the old Penguin.

[Simon_H]

@aristotle Why? If Panda punishes thin content-type issues, too many ads on the page that aren't clearly separated from content, etc, that's surely less likely to be site owners intentionally trying to manipulate the SERPs than link building.

[aristotle]

Simon_H -- I'm looking at it from Google's perspective. Panda improves the search results a lot more than the old Penguin does (did).

[Simon_H]

@aristotle Ah - fair point. Panda issues affect what the user actually sees, Penguin issues are totally unseen by the user. Still, I'd be interested to know if Penguin 4 is a sign that Google is pulling back on the punitive side of things. Panda could still filter chunks of a site that it felt were low quality without the site-wide punitive side. In fact, Gary has in the past suggested noindexing site pages that are low quality and cannot be thickened up to prevent Panda issues. The ideal situation is that people don't need to worry about that, and Panda would also work on a truly granular level and simply filter the 'thin' pages without killing the good pages.

Anyway, sorry, enough about Panda. This is a Penguin thread!

[Balle]

According to Gary Illyes (Google) on Twitter the part of the update that removes penalties from Penguin 1 to 3 is currently being rolled out now and over the next few days.

As far as I understand this should make it possible to recover - OR be hit by once more but now by the new Penguin 4.

[EditorialGuy]

Awesome, this means that we can start buying/selling links again.

Sure, if you aren't worried about manual penalties.

[Robert Charlton]

Regarding Penguin recoveries... Gary Illyes in ongoing discussion with Barry at SERoundtable, has clarified that the algo has been rolling out in phases, and that the phase in which demotions will be removed started late yesterday afternoon, ie Sept 28, 2016. This phase should continue for the next "few days".

Keep in mind that spammy links have been devalued, so sites will may not appear to have "recovered" unless they've gained non-spammy links in the interim....

Google: Penguin Recoveries Rolling Out Now Over Next Few Days
Sep 28, 2016
[seroundtable.com...]

Mod's note: For now, we're keeping all Penguin rollout discussion on this thread.

[aristotle]

Evidently Google has given up on its original vision for Penguin. After more than four years of trying to make it work properly, they've abandoned their original approach and turned to something quite different. I suspect that Penguin will be much less of a factor from now on, and with less impact on individual sites.

[Nutterum]

Not seeing any recovery of penguinized sites in the top 50 SERPs for a big travel related cluster of keywords and phrases. Will wait a week and report back.

[redgorillas]

I just got fresh Semrush tracking report. So we started to see recovery in USA for our site selling software. It was hit several times by Penguin. We got many our long tail keywords getting +10 + 20 bonus and many of them now visible on first page. Popular keywords movement is minimal. So average site visibility improved around 20%. We still see old data set in Europe and other countries. Frankly speaking we can see 3 data sets. One where all of our keywords are punished and we have just 10 keywords on page 1. Second is a bit better and more long tail keywords are on page 2 and in USA we can see best one - around 60 target long teil keywords now appear on page 1. Traffic still almost the same.

Really hope that Google will not put old penalites back for our site. We were working on our sites for 2 years.

[EditorialGuy]

Evidently Google has given up on its original vision for Penguin. After more than four years of trying to make it work properly, they've abandoned their original approach and turned to something quite different.

It would be odd if Google's approach hadn't evolved over four years' time.

[Simon_H]

There certainly seem to be some examples of recovery, but I don't think we're even remotely close to all of P3 being lifted based on what people are reporting. Gary has never been great at time estimates. When he said 'over the next few days' for P3 being removed, I'm guessing that's about as accurate as him saying that he expected P4 to be released in 2015.

People say that sites won't necessarily see a ranking increase when P3 is lifted if their spam links were driving their original great rankings, but that's not necessarily true. Those spam links should already be passing negative pagerank under P3, so removing P3 would theoretically switch them to zero pagerank. Hence, rankings should still increase to some extent. And if any P3 sites have accumulated any 'good' links over the past 2 years - which they very likely will have done - the traffic increase should be noticeable. So this seems more that P3 is a long way away from being lifted, rather than it's been lifted and most sites haven't been impacted.

[Simon_H]

Something else that is confusing about what Gary has said... How can P3 be removed from a site if P4 hasn't yet been applied? Let's say Google simply removes P3. That would mean all those spam links will suddenly have equity. So the assumption is that Google isn't actually just removing P3, it's also negating all of those spam links at the same time. But that's the job of P4 and happens when those spam links are crawled/indexed, and it's certainly going to take more than the next few days to crawl/index most of the web! So this doesn't make a whole lot of sense.

I'd be interested to know what sites already seeing a recovery have in common. For example, are they sites of a certain size, or use the disavow in a certain way, etc.

[aristotle]

How can P3 be removed from a site if P4 hasn't yet been applied?

Perhaps P3 won't be lifted from a site until after it has been evaluated by P4

[NickMNS]

@Simon I believe that what Gary is saying is all smoke and mirrors. What ever changes, if any, have already been done. Any changes moving forward will be statistically insignificant. That is they will occur but be small enough in scope and/or scale such that they will be indistinguishable from the normal day to day fluctuation. This will keep webmasters in a state of constant paranoia always wondering if a sudden drop in traffic is the result of Penguin, but without having any means of knowing with any degree of certainty.

I think that Penguin/Panda was a massive mistake on the part of Google. Not in its intent but in its execution. Now they are trying to undo the damage done. Eventually in the coming months and years, this will be forgotten and all parties will be better off.

[Simon_H]

@aristotle Exactly. But that would surely take months if it relies on all links to P3 sites being recrawled and evaluated. Not just a few days.

@NickMNS Gary has made a clear and unambiguous statement about P3 being removed from all sites. It can't really be misinterpreted and so that's going to be a very difficult one for Google to come back and say "ah, actually, no, we didn't mean that". And any argument that some P3 hit sites have had P3 removed but didn't notice the difference wouldn't hold for the reason I made in my first comment above. Perhaps Google will suggest that some sites aren't seeing a change because they weren't actually hit by Penguin, but if that site saw a sudden massive drop in traffic around a previous Penguin rollout date (which is how most determine whether they're under Penguin), that argument wouldn't apply either. So very interested to see if/by when P3 will be completely removed.

[NickMNS]

My point, and this is pure speculation so I may end up being wrong, is not that any given site will see some relief from a P3 penalty. What I am saying is that relief may occur in small increments. Increments that would be small enough that they would not be detect, when they occur. Say a 10% increase in traffic on a given day. The webmaster would be pleased, but wouldn't necessarily think it was P3 being partially lifted since any other day you would +/- 5% fluctuation in traffic. If this continues over several month, the webmaster would see a net increase traffic over time, but when, what and where the P3 penalty was lifted will be unclear. Add to this site changes and other factors, it will become impossible to tell what is what.

Personally, if I had to design such an algo, this is the strategy I would apply.

Again, I am not saying that they are lying out right, P3 will be lifted. But things will not be so cut and dry as they have been in the past.

[glakes]

@NickMNS

All the rank trackers are still hot so there is still hope for those expecting a Penguin recovery.

I think that Penguin/Panda was a massive mistake on the part of Google.

Completely agree. Disavowing links and sending DMCA takedown requests, because of duplicate content created from scrapers, is a large burden for individual small business owners. For a company that employs the brightest people, I would think they could have gone about it in a better way that would not have created an environment that cost small business owners billions of dollars in lost time and services related to recovering from the animal penalties.

[dipper]

okay .. so, I think it's been missed - Penguin really only last 4 years, and it's kind of irrelevant now. Google have reverted to what it was before penguin .. links are either positive, or nothing (with regards to Penguin). Matt Cutts war on webmasters is over.

Also - has anyone else seen partial recovery? - as in, mobile jumped up visibility, but desktop remained stale?

[Shai]

Interesting observation historically associated with Penguin updates:

Searching for a phrase with a slight spelling mistake brings up completely different positions for some sites even thought Google recognised the error and claims to be searching for the correct phrase.

Try it if your site is seeing movement.

[Peter07]

Links are either positive, or nothing: so it is not possible to sabotage the competition anymore with bad links. Finally, this is how it should be. ;-)

[EditorialGuy]

links are either positive, or nothing (with regards to Penguin)

Isn't it likely that Penguin looks at linking patterns, not just at individual links?

[Shepherd]

Isn't it likely that Penguin looks at linking patterns, not just at individual links?

Interesting thought, likely, don't know, possible, yes.

If that is the case, is it your thinking that google might discount all of a site's links if there is a significant pattern of "bad" links?

[dipper]

@Peter07 - It seems that way - well, not with Penguin anyway, and like it was 4+ years ago, but you can still have manual penalties.

@EditorialGuy .. I preface this with a imho - yes, patterns likely do matter. I don't think like @Sheperd that bad links are ignored, they are just irrelevant. How do you think Google look at links in patterns?

Another question I have. Why did Google, and don't Google now, need the disavow file? .. according to Gary Illyes he hasn't ever seen a real negative SEO result, and he said Google haven't altered Penguin mechanics (beyond going real-time).

[aristotle]

Most likely Google frowns on it, but their hands are tied as far as taking any action.

[EditorialGuy]

If that is the case, is it your thinking that google might discount all of a site's links if there is a significant pattern of "bad" links?

Beats me. I'd be curious to know what's bad enough to invoke Penguin but not bad enough to invite a manual penalty.