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.

[SnowMan68]

Well that was underwhelming. WTF

[NickMNS]

@SnowMan68 were you really expecting a big shake up like in the passed? The only thing this announcement confirms is that there will no longer be any big shake ups in the serps.

[SnowMan68]

@NickMNS

I was expecting something...haha. So far...pretty much nothing.

[Barbados]

"The only thing this announcement confirms is that there will no longer be any big shake ups in the serps."

Are you sure?

[SnowMan68]

previous moves people saw could be testing. I suspect we are going to see moves in the next couple weeks.

[engine]

At last, Google is prepared to confirm something that we had a pretty good idea about.

Penguin 4.0 is the live version.

[martinibuster]

The only thing this announcement confirms is that there will no longer be any big shake ups in the serps.

Maybe. Maybe not. I believe there will continue to be shakeups because nothing is perfect, not even Google. The core algorithm cannot identify all forms of spam. That's why they have to take "manual" action to remove the spam that "automated" actions failed to catch.

After the manual action they may "refresh" the algorithm in order to better catch what the core algorithm overlooked. Refreshing the algorithm is a euphemism for identifying the "fingerprints" (classifiers) of the uncaught spam and using that to train the algorithm to catch that spam.

It also means we're not going to comment on future refreshes.

That statement confirms that Google will continue to update ("refresh") the algorithm and this will naturally shake things up now and then. But apparently they're not going to talk about it.

Do you think they'll stick to that promise?

[EditorialGuy]

That statement confirms that Google will continue to update ("refresh") the algorithm and this will naturally shake things up now and then.

Going forward, shouldn't much of Google's updating the result of machine learning? And if that's the case, wouldn't it seem likely that most updating will consist of incremental changes rather than the major "update" events of the past?

[NickMNS]

My point is not that serps will never be shaken up.

What I mean is that future shake ups, including to some extent the one we are currently in the midst of, will be limited to a relatively small subset of websites, in such a matter that it will be difficult to know for sure whether any changes experienced are the result of a an algo change or just the result of normal turbulence.

When Penguin 3 rolled out it was pretty obvious that it was in fact Penguin. Many sites with link issues were penalized all at once.

Now look at the current situation, that is the increased turbulence since the beginning of the month. Several people here are seeing changes, for others (myself included) it is business as usual. No one up until this announcement was able to say for sure what happened. Even post announcement we are unsure whether the turbulence we have seen thus far is Penguin, or if the impact is still to come.

It is impossible to know for sure.

Google's goal all along was to confuse webmasters such that they are unable to gather any useful information that would allow them to game the system. So congratulations Google, that goal has been achieved.

My view is that the Penguin/Panda roll out should have been done in this fashion from day 1. It wasn't, and that was a big mistake on Google's part.

[smilie]

>> @NickMNS: Google's goal all along was to confuse webmasters such that they are unable to gather any useful information that would allow them to [s]game the system[/s] make any money from their websites and their content.

Here, corrected.

Amazon is now 30% of all US online commerce and growing.

Incidentally, Amazon is also part of coalition of big merchants for taxing internet transactions. If it isn't obvious why , it is TO FINALLY KILL SMALL ONLINE MERCHANTS once and for all. IMHO Obamacare tax was part of that.

So the can grow from 30% to 50% of all US online commerce. I thought internet was supposed to level field for everyone, not just Google, Amazon and Walmart.

By making FUD and keep obfuscating things, Google is clearly a part of that big merchants coalition to milk online for billions. And Google is clearly against small merchants and FOR big business as you can't operate a small online business in Google's FUDed waters anymore.

The only reasonable way to let this go was to basically either cancel Penguin altogether , or make it a LOT lighter. But the greed knows no bounds.

A big market crash is coming, it is imminent. After all, small businesses (and not big) used to make over 50% of all jobs. And online commerce was the only growing niche in US for years now , and mom-and-pops are now choked by animal updates to near death.

[edited by: smilie at 3:18 pm (utc) on Sep 23, 2016]

[martinibuster]

if that's the case, wouldn't it seem likely that most updating will consist of incremental changes rather than the major "update" events of the past?

future shake ups, including to some extent the one we are currently in the midst of, will be limited to a relatively small subset of websites, in such a matter that it will be difficult to know for sure whether any changes experienced are the result of a an algo change or just the result of normal turbulence.

You may be observing trees. Step back.

Allo: Welcome to the forest
Allo features an AI assistant that people all over the world will be talking to. Right now it's clunky but as you continue to use the AI assistant it learns what you want and improves. This is search tailored straight to the individual. The collected data can be used to create a better algorithm that will understand the nuances of what is desired and required, specific to you as an individual.

From NYTimes
Allo’s Tryout: 5 Days With Google’s Annoying Office Intern [nytimes.com]

To understand how Allo works, it’s easiest to think of the app’s A.I. assistant as an office intern who is lurking in the background, eager to chime in. The assistant analyzes messages you have typed or dictated and, when appropriate, springs into action with automatically generated phrases you can choose to reply with or suggestions for Google searches that may help accomplish tasks.

Google said it was still improving and refining its algorithms, and Allo’s assistant will get better over time.

Once Allo’s assistant matures, the Hangouts app will become redundant and you’ll be able to delete it from your device. The catch, of course, is that Allo’s A.I. won’t become sophisticated until more people use it and share feedback.

Penguin and this current algorithm are only temporary, perhaps for the next couple years? Hard to predict since it is dependent on Android/Apple users and in the near future the people who still use the desktop. The AI behind Allo is the future of search.

[glakes]

So the can grow from 30% to 50% of all US online commerce. I thought internet was supposed to level field for everyone, not just Google, Amazon and Walmart.

I do get some satisfaction selling on Amazon when new customers visit my site after their Amazon purchase and ask why prices on my site are less. I tell them that Amazon charges $40 a month for Pro seller accounts, and then takes 15% off of the total sales (including shipping) as their fee, which must be paid for by the customer because I'm not in manufacturing to build and give stuff away for free. If more Amazon sellers did the same, it might help to gradually educate consumers that buying products in big marketplaces often cost more. Educating consumers is about the only way the playing field will get leveled because no government agency is going to step in and the 800 LB. gorillas are going to keep consuming as much of the economy as they can.

[TechNoob]

I don't personally think anything dramatic will take place in future algorithm updates either. What could truly shake things up? Stricter backlink policing?

Maybe fact checking.

It's 2016. At this point, I only see the mobile SEO arena potentially getting shakeups in the future.

[EditorialGuy]

Google's goal all along was to confuse webmasters such that they are unable to gather any useful information that would allow them to game the system.

I think it's more likely that any lack of "useful information" for gaming the system is a side effect of Google's use of AI, personalization, etc., not the reason for Google's advancing beyond the relatively simple search algorithms of a decade ago.

To use an analogy, it's like the invention of the tubeless tire: The goal was to make a better tire, not to put tube makers out of business.

[Balle]

I don't see much change at all.

In the past a penguin refresh/update announcement used to be accompanied with something like "This update affected 1%" of web searches" or something like that.

.... hopefully it's still rolling out.

[dipper]

is that like Panda "part of the core algorithm" which still needs someone to push a button to start the refresh?

"granular" - page level, instead of site-wide?

Edit: seems Jennifer has answered these now > [thesempost.com...]

[austtr]

So if Penguin is now fully absorbed into the core algo because it can recognise spam indicators all by itself, then presumably disavowal files no longer serve any purpose.

Can all previously submitted disavowal files be "reset" to empty plus there is no longer any need to submit new disavowal files.

Yes or no.

[Walt Hartwell]

Can all previously submitted disavowal files be "reset" to empty plus there is no longer any need to submit new disavowal files.

I do not know how Google would treat an empty disavow file, so I downloaded my last disavow, removed everything except one foreign language domain and uploaded my one domain disavow.

I have not seen any noticeable differences that I could attribute to the existence, or effective non-existence of a disavow file.

[FranticFish]

any lack of "useful information" for gaming the system is a side effect

We have had 'not provided' for years. Do you think Google just deletes this data or keeps it for themselves?
And recently, Google took up to 75% of previously shown keywords in GSC and lumped them together as 'not set'. Again, do you think they delete this data or keep it for themselves?

Did either of those 'make things better'? For who? Oh, and before anyone says 'privacy', remember that you can run keyword targeted ads in GMail messages.

[Shai]

So if Penguin is now fully absorbed into the core algo because it can recognise spam indicators all by itself, then presumably disavowal files no longer serve any purpose.

I don't think anyone including Google are claiming their algorithms are 100% perfect all the time so therefore the need for a disavow file remains for collateral damage and also other cases such as Negative SEO attempts and less knowledgable business owners hiring the wrong type of SEO agencies/people and in turn those completely ruin the link profile of the site hence needing a way out via the disavow file.

[Shai]

This is also really interesting from Jennifer's post linked to by Dipper above: It can be quite misleading so a careful read is recommended.

“We look at the webmaster’s page and devalue incoming spam as we recrawl and reindex the particular page. So it’s the webmaster’s page that we’re taking about, not the spammy site.”

So a page’s rankings will be reevaluated every time Google recrawls the page, and the impact of any incoming links – and spam – will be adjusted immediately.

Now, this still means that Google needs to crawl those pages where the spam links are, so that a disavow will be processed or for Googlebot to see that the link is no longer on the page. But now those rankings can be adjusted in real-time, rather than waiting for a push.

This means that for the change to take effect, this is the order of things:

1. Disavow file in place with the incoming link/domain disavowed.
2. The page the link is actually on needs to be crawled/discovered.
3. the page that the link targets is then crawled but this time without the effect of the link.

[engine]

A key factor of all this is elevated crawling.
i wonder if we're going to see sessions of elevated crawling, or whether googlebot is going to settle down and crawl more frequently on a regular basis. We've yet to see.

[EditorialGuy]

We have had 'not provided' for years. Do you think Google just deletes this data or keeps it for themselves?

I have no idea, but the notion that personalization, the use of artificial intelligence, etc. were introduced to confuse SEOs and site owners (as was suggested earlier) strikes me as being awfully farfetched.

[EditorialGuy]

i wonder if we're going to see sessions of elevated crawling, or whether googlebot is going to settle down and crawl more frequently on a regular basis.

More intense crawling would be worthwhile--especially in terms of discovering and indexing new "inside pages"--but I can't see Penguin being the driving force behind it. After all, Google didn't seem to be in any great hurry to get Penguin 4.0 out the door.

[martinibuster]

...the notion that personalization, the use of artificial intelligence, etc. were introduced to confuse SEOs and site owners (as was suggested earlier)

Whoa, whoa, whoa... I never said that and most importantly that's not at all what I meant. :o

Allow me to clarify please
I introduced the point about AI in response to the sentiment that this algorithm is it and that future changes will be obfuscated. The intent of my post is to show you that Google is taking it's very first steps toward a new (possibly link-less) algorithm right under your noses. Ray Kurzweil is working at Google to develop a 100% AI to handle search. Allo is the first baby step of an AI system to replace search as we know it.

Nobody on any other forum, news site or white board has noticed the implication of Allo or made the connection. The point of making this observation is that if you think Penguin is interesting or bad or good, take a look at what's happening over there with Allo because that is the very first introduction of what is going to come.

[IanCP]

@glakes

I tell them that Amazon charges $40 a month for Pro seller accounts, and then takes 15% off of the total sales (including shipping) as their fee, which must be paid for by the customer

Some of that 15% charged, after shipping costs, is very often shared with Amazon affiliates who provide a real service. Rates vary for affiliates but 4% to 8% would be typical.

Amazon affiliates provide a real service because most of us have quality content sites, where we discuss all kinds of widgets, the ins and outs of green widgets versus blue widgets, with a direct link to a specific widget on Amazon.

That is called pre-selling.

At Amazon, the folks in the Amazon Sellers programme often not only compete with Amazon itself, but also compete with other sellers.

Amazon does not always retain all of that 15% - after shipping costs.

It has often been claimed over the years that Amazon now - having established themselves on the back of affiliates [I've been an active one for nearly 20 years] - must no longer need affiliates, given their behemoth size.

The fact that Amazon still supports their affiliate network suggests otherwise.

Well @glakes, what percentage of your sales on Amazon come from those affiliates? Sales perhaps you might not otherwise have made, sales which might have gone to a competitor?

[EditorialGuy]

Whoa, whoa, whoa... I never said that and most importantly that's not at all what I meant. :o

Not to worry, martinibuster. I was responding to earlier comments by members. I don't regard you as a purveyor of conspiracy theories. :-}

[goodroi]

And now back to our regularly scheduled topic - Penguin

Let's stay focused on this topic and if you want to talk about Amazon, Google finances or conspiracy theories like Google covering up the Roswell incident ;) please start a new thread in the correct forum.

[lightnb]

I'd really like to know if anyone at all has seen a recovery or a hit from this update. our traffic got crappier about a week before the launch announcement, but keyword position didn't really change, just the normal +/- from week to week. After the announcement, no change in Google traffic (still crappy) and no change in ranks despite disavowing hundreds of links.

So either Penguin 4 is a placebo, or it will take a long time to roll out, or all the spammy links pointing at us didn't really drop/affect our ranking. But then, why would we rank in the 50-70's for many high-volume commercial keywords where we rank top five for them on Bing, and there are only about 5 competitors? What else could cause unnaturally low rankings of specific words when we have a quality site full of unique content but 100's of spammy links thanks to a competitor? Penguin seems the most obvious answer... and yet, nothing.

So if you have a site that was previously Penguin hit (or presumed to be) could you share what you're seeing?