Theory on Panda/Penguin false positives

[diberry]

There are always claims of false positives when Google does something with the algo. A lot of these claims are incorrect, the result of webmasters not looking at their sites harshly enough to see what the algo is seeing. I've never been convinced there were any false positives, until Panda hit people I know, whose sites were extremely unique and creative, nothing like "content farms." But they were, we agreed, perhaps a bit "over-optimized" - targeting hot keyphrases and so on, which we somehow thought might be part of Panda.

But supposedly that's not part of Panda, because now we know it's part of Penguin. Right? Or is that right?

When Panda came out, and we heard is was about thin content, I knew one of my sites was vulnerable. I was working slowly to improve the content, but my focus was on other sites so I was just waiting for Panda to catch up and hurt me. I had never been any good at link building, so all my links on all my sites are natural. On this particular site, my backlink profile was really weak because in that niche, people are very stingy about giving out free, natural links. I thought that might get it in trouble with Panda, too.

To my shock, Panda left my site alone, but Penguin got me - on my least optimized site. Yes, I realize I'm self-reporting and you have the right to be skeptical.

One possibility: my Penguined site is over 6 years old, so I have the sort of unsolicited spam links everyone accumulates over time, like updowner. I only have a handful of quality links, because of my lack of skills at link building of the white or black hat variety. Could Penguin be mistaking weak profiles on older sites for spammy profiles?

Or are Penguin and Panda intersecting and overlapping in ways we still don't understand? Why did people with unique and interesting content get Panda slapped, and why did my least optimized site get Penguin?

[edited by: tedster at 9:35 pm (utc) on Jul 3, 2012]

[varun21]

Matts B**** Cutts will be so happy to read all these threads trying to reverse engineer their algo which rests on 200+ "signals".

[driller41]

What they all have in common: Maybe less than 20% of their entire site content could be considered truly fresh or truly unique. It's mainly a regurgitation of what's already on the web.

I believe this is what Panda would a have been targeted at, remember Panguin is 'spam' penalty. I am wondering what google considers spam these days - onsite duplication is not really spam.

[diberry]

driller41, I agree and have always said a Panda slap would've made sense on my site. But I got Penguin instead, and no one has yet put their finger on why. But there's a thin line between the rehashinh content because you're just not that good a writer and rehashing content in hopes of ranking. If the algo thinks you're doing the latter, I find it plausible that it might give you Penguin problems.

[Sand]

I find it plausible that it might give you Penguin problems.

I don't have any hard evidence yet, but I have a strong suspicion that at least a part of Penguin is set on determining what the *intent* of a site is.

I say this, because I had a couple ancillary sites that had squeaky clean link profiles, but honestly served no greater purpose than to send traffic from one site to another. They were (rightfully) knocked down a few levels with the first Penguin update, and I would surprised if they ever recover.

In the name of science, I removed all the links to the 'parent' to see what happens.

[varun21]

I'm not sure how they can calculate the "intent" of my website. It's a very genuine photography (don't know if I'm violating any rule here) blog. It's got many images and everything original. The only thing I ever did was hire an SEO firm for directory submission (about 1k-2k links). But I did get some genuine links from ehow etc. Penguin killed 80% of my traffic and no level of optimization will help now. I'm moving to a new domain without a 301 redirect. I'm going to have to load a page saying this "page" has moved, please fix your bookmarks and you'll be redirected in x seconds. Seeing that there are no more penguin updates is very discouraging for sites that are trying to recover. But I bet there will be at least one before christmas (cause everyone including Google doesn't want to miss money then). Pls tell me penguin is all about external links.

[varun21]

Additionally, the reason (I feel) Penguin is link based is because (since) you can't take off external links immediately, Penguin doesn't run that very too often. It's allowing us time to clean up our link profile and then may be after 3-6 months it will run again.

[Robert Charlton]

(about 1k-2k links)

varun21 - I'll bet that's your problem. That many coordinated links, even if the sources were OK, certainly must jump out on your link graph. The "intent" that Google is looking at isn't the intent of the site, btw... it's the manipulative intent behind those links.

As I think about it... much of the SEO I've done for some clients, well before all the Animals and MayDay, all the way back to the Florida update, has been either to clean up what other SEOs have done, and/or to protect these clients from themselves.

[Robert Charlton]

It's allowing us time to clean up our link profile and then may be after 3-6 months it will run again.

I agree with this observation, but I think there are several things going on here. In addition to allowing time for links to be taken off, Google is most likely also looking at rate of acquisition of new links... and it will be much easier to do because Google will have a well-defined group of sites to watch.

[varun21]

Thanks Robert. I'm pretty naive in the SEO world but my back is against the wall since I got hit by Penguin (and I guess by Panda too in a very limited way just a few days before it).

Never imagined about the "acquisition" of new links though. I understand we can't be playing "God's mind", but rate of acquiring new links could be it. And I bet Penguin is going to run again; at least once before Christmas - G***** has to get it's act together before it can make the most of the season.

[diberry]

Varun21, I don't think Penguin can be just about backlinks, because my backlink profile is weak rather than spammy, but I agree with Robert Charlton that in your case, it's likely to be the backlinks.

As for how Google would determine your intent... well, obviously they can't read minds, but I can imagine them feeling they have a pretty good "profile" of how an overly or wrongly optimized site appears. I.E., the backlinks look like this, the bounce rate is like that, etc. (I also suspect they would have lots of different profiles for differnt types of queries.) They might be ranking sites against such a profile, and if your site fits it, say, 85%, then they slap you.

Just to keep us from comparing apples to oranges, my site was not penalized according to Google. I lost about 80% of Google traffic, like you, but it was just due to changes in SERPS that happened on the day of Penguin. I think it's helpful to remember that Penguin is not a penalty, it's just an algorithm, and it affected a lot of people who did not have penalties or "unnatural link" notices.

[Robert Charlton]

I lost about 80% of Google traffic, like you, but it was just due to changes in SERPS that happened on the day of Penguin.

Something I've mentioned in a bunch of threads but maybe not this one, that's very likely due to upstream link sources getting hit. (This is another reason, btw, why it's good to have very diversified inbound links).

[diberry]

I remember you saying that. I wasn't tracking the people who link to me, because I wasn't doing any SEO, so I can't be sure, but this may well apply to my site. (Though, the site that most often links to me quite naturally, within the text of pages, is doing better all the time.)

And if you're right - and it makes total sense to me - then a site can be affected badly by Penguin without actually having done anything spammy. A lot of my backlinks for this site, for example, are from little blogger or wordpress personal blogs, putting my site in the sidebar because they like it. If those got hugely devalued during Penguin, it might expect why the rankings just sort of fell out from under my whole site in general. (That's just one possibility, for illustration purposes.)

Another big source of traffic for that site - and not for my others - is Pinterest, and it's mostly all pointing to one page that I guess just really grabs Pinterest readers. I didn't write it for that purpose, but maybe Penguin thinks it's linkbait.

There are just SO many possibilities, I don't think it's worth trying to figure out the riddle. I'm devoting my time to improving the existing content, adding new content that I believe is truly unique, and reaching out through social media to find visitors and subscribers. If this creates the signals Google is looking for, then my rankings will improve. If not, I'll still be increasing traffic through other streams, and that's what a website needs.

[Robert Charlton]

Pinterest

I can't imagine that Pinterest as a very solid or steady source of inbound links... at least of the kind that would support long term rankings in Google. These aren't what I'd call editorial votes.

They might well drive traffic for a while and act to confirm inbound link acquisition, but IMO probably won't help you build authoritate links. I've got to think that social links like this are going to fade considerably faster than blog links... and if your only source of inbounds was from Pinterest, then you have a lot of work to do.

Take a look at this thread, pre-Pinterest, but still possibly food for thought...

Does Google "Age" Your Backlinks?
http://www.webmasterworld.com/google/4176006.htm [webmasterworld.com]

[diberry]

I agree, Robert. My backlinks were mainly a mix of small blogs putting my site in their sidebars, a couple of bigger, more authoritative blogs linking to me very naturally in their page text, and social inbounds. I never thought the social inbounds counted as much for "links" because there's been debate about whether Google even considers them - but they do bring traffic, and at the end of the day, that's what I'm after from Google or anybody else.

That's why I've said my backlinks were weak, but not spammy. I know there's a lot of work to be done and I've been saying that all along. :) My only point with this thread is that it's erroneous to think being affected by Penguin means Google thinks you're spamming. I think either Penguin is broader than Google's "anti-SEO" statements indicate (they're not required to tell the whole truth), or there's collateral fallout as you suggested with devalued backlinks, or... something.

Most of my links actually were pretty old. While I do suspect Google might weigh the age of a link in its factors, I also think I haven't gotten many new links in the past year and THAT could be hurting me.

I doubt I'll ever know why I got the Penguin. What I do know is that since *I* know I wasn't spamming, it would be silly to try to "fix" something that might have been perceived as spamming. I just need to work on getting more loyal visitors, and hopefully at some point the algo will catch the right signals.

[diberry]

I've had a thought. Maybe it should be a new thread, but it does tie into the idea of perceived "false positives."

A lot of people badly hurt by Penguin have "no manual action" emails, post-Penguin. If Penguin is mainly about spammy backlinks, you would expect affected sites to have gotten "unnatural link" notifications. Possible explanations:

--Penguin is about a lot more than that, maybe more subtle forms of spam or even any hint of SEO of any kind
--Some affected sites are hurt because their backlinking sites got hurt
--Anything else?

In my own case, I don't think the second option could be the whole explanation. I have a handful of strong sites giving me quality inbounds, and then a bunch of little blogs plonking me in their blogroll - that kind of thing. It's possible the little blogs - most of which are on Wordpress.com or Blogger - got devalued, but that wouldn't seem to explain my 80% drop.

So in a way, I'm returning to my original theory in this thread: Penguin is about a lot more than backlinks. In fact, I think Penguin may represent a whole new way of looking at spam/SEO, just as Panda represented a new way of evaluating quality. I.E., a few years ago, you were either spam or you weren't. Then Panda came and you were either spam, lower quality content or better quality content. What is Penguin trying to sort out? Could it be trying to detect and punish ANY form of SEO - that is, even though Google said "aggressive spam" which sounds like "the worst blackhat spam" in my mind, what they really meant was "any attempt to do well by our algo at all whatsoever"?

And even then... I wasn't doing ANY SEO, so I still just don't get Penguin, or don't buy what Google has told us about its intent.

[gouri]

I have read that Penguin involves backlinks but that there could be more to this algorithm than that.

I am wondering if you guys think that backlinks are a component of the Panda algorithm along with other factors or is this only a part of Penguin?

[indyank]

diberry, on reading your story I do get the feeling that your site's main trouble is with those extenal links to it.

1) a mix of small blogs putting my site in their sidebars - this would have definitely landed you in trouble, more so if they weren't branded links.
2) Blogs, except the high profile ones, have in general been devalued a lot by the Google animals. So, if your main source of links were blogs and a few or most of them were victims of the Google animals, your site would have been definitely affected indirectly as well.

I am wondering if you guys think that backlinks are a component of the Panda algorithm along with other factors or is this only a part of Penguin?

As I have already mentioned elsewhere on webmasterworld, this is my take. Panda is definitely about links and their characteristics too and not just about content. Penguin is predominantly about backlinks and to a small extent about spam (not just low quality) content.

[indyank]

A lot of people badly hurt by Penguin have "no manual action" emails, post-Penguin.

I think it would be interesting to hear whether any Penguin victim received this recent "unnatural links" notice at all...

[varun21]

Penguin victim here. Waiting for a notice everyday, everyminute. At least anything that can tell why my site was penalised. No notice so far ever since my site was penguinised in the last week of April.

[indyank]

diberry, Have all those external sidebar links from blogs been removed? I do know that you mentioned none of them were solicited links but were natural.Unfortunately, Google animals would definitely not be able to know it.

I also feel that you might not have received any real mileage from those blog sidebar links even prior to the release of Google animals but would have definitely attracted "site wide action" (as distinguished from "targeted action") from the animals. IMO, a "site wide action" is definitely penalizing in nature. What else could it be?

[indyank]

Penguin victim here. Waiting for a notice everyday, everyminute. At least anything that can tell why my site was penalised. No notice so far ever since my site was penguinised in the last week of April.

Yes, this is what I expect to hear from Penguin victims. It would be interesting to hear otherwise.

[indyank]

At around the same time Panda versions were released, Google also brought about a major change in their algorithm to offer diverse results for queries, as explained in this.

Eric Enge: Going back to the way we started this chat, Google has compelling reasons for offering diverse results. Understanding this could offer new online businesses a way in the door. In the past people have referred to this as Query Deserves Diversity.
Matt Cutts: Yes, that is a part of what our algorithm does: work to find quality diverse results that help solve problems for users. I would discourage people from thinking about from an algorithmic perspective though. What they should focus on is looking at the overall landscape of their market.

So, some Panda victims would never be able to get back ranking for their affected pages or sites, unless the pages offer "diversity" for the query they used to rank for.However, they might get back the power to rank their new pages if they offered diverse and unique content for the targeted queries.

[diberry]

Indyank, I am not penalized. I've gotten the post-Penguin "no manual action" email.

Now, as for the sidebar links, I don't think we're talking about the same thing. These aren't blogs for profit - these are the 2010s equivalent of the 1990s homepage: people blogging for family or just for fun. They tend to put all their favorite sites into the blogroll, like fans. They've never heard of SEO, they're not running ads. Every blog I run or follow attracts these sorts of inbounds, but only this one got Penguinized.

I think for these to be a problem, there would need to be some indication that the little blogs were fake. Say, virtually the same blogroll on each one, or the topics of the little blogs always being tightly matched to mine. That's just not the case. It's just random people putting up blogrolls for fun. I can't see Google punishing webmasters because non-webmasters blogging for free decide to plonk them in the blogroll.

[indyank]

Indyank, I am not penalized. I've gotten the post-Penguin "no manual action" email.

I am also saying that you were affected by Penguin but for those backlinks which you described and Google misinterprets as planted or solicited links. To treat Penguin as a penalty or not is up to you and if you still feel you weren't affected by the backlink evaluation part of the penguin algo, you could continue to explore other reasons as you are the best judge.

I am not sure what you mean by the "no manual action" email. Are you talking about Google's response to your reconsideration request? "no manual action" email is sent only in response to such requests. But I am talking about the recent "unnatural links" warning notification that google sends out for individual suspect pages on the site.

[diberry]

Yes, I asked for reinclusion and got that response, and I have also never gotten an unnatural links notice on any site/page.

I run quite a few sites, and monitor some more, and they all have "homepage links". This is the only one affected by Penguin, so at the very least there has to be more to it than just the "homepage links".

[diberry]

Okay, since the Panda update on the 29th, I've seen a steady rise in traffic. It was hard to tell at first whether this was related to Penguin, since I've made a lot of other changes which took the rankings even LOWER than Penguin did.

But today I'm seeing a lot of old inbound referrals that I haven't seen much in months or even years. I'm wondering if those sites have just recovered from Panda, and they're getting more traffic to send me, and also their inbounds might now have more value for my site.

I think this supports what you've been saying, Robert_Charlton, about being affected by backlinks even if none of them are particularly "spammy." If enough of them have been affected even mildly by Penguin OR Panda, the cumulative effects could impact your site deeply without you having done anything "spammy" or even SEOy.

[diberry]

Okay, this: [webmasterworld.com...]

From the beginning, I've wanted to suggest - and sort of hinted - that the reason my site got Penguinized was that I had responded to Panda by making changes. Now I'm convinced: if Google is that focused on punishing SEOs, then it's plausible to me that they would release an algo that's overzealous in its attempt to identify them.

Just a quick summary of the facts:

I run four sites, all built for users, not Google. Only one got hit. I've never gotten any "unnatural link" notices, and it's confirmed the Penguinized site was not penalized. Let's call it Site C.

Back when Panda hit, Site C was NOT hurt, but I figured it was only a matter of time. I began making changes based on my perception of (a) what users want and (b) what Panda wanted.

Did Penguin slap me merely because I made changes in response to Panda? I'm sure some will disagree, but that's what I've been thinking for a long time, and as far as I'm concerned, this confirms it.

[Robert Charlton]

diberry - One another thread [webmasterworld.com...] you just posted...

I've been thinking more about 1script's remarks, and realized, duh, I did make one tech change around the 23rd.

Basically, I had changed my domain name months ago for branding purposes from OldName.com to NewName.com and 301'd everything from OldName. Around the 23rd, I switched OldName to a reseller hosting package, since all it's doing now is redirecting inbound links to OldName that I can't get people to update. There was no downtime for either site, because I left OldName's old hosting up for four days to cover the propagation period and then some. I've changed hosts on my main sites many times in the past and NEVER had a Google traffic drop like this, so it's hard to believe changing hosts on the redirected site would have such an impact.

I just want to clarify, is the site you're discussing in this thread the one referred to in the quote above? If so, might this perhaps open or re-open an avenue of inquiry that we might have missed or glossed over?

Changing a domain name in close proximity to making lots of changes on a site is something that Google has frowned upon for a great many years, probably going back all the way to the Florida update, at least. This was true even before the patent for spam detection also under discussion.

[diberry]

It is the same site, but I changed the domain name in late May, after Penguin had decimated my traffic. There had been no significant hosting, tech or site layout changes for probably at least a year before Penguin hit me. I had actually wanted to change the domain name for years, but didn't want to risk hurting my Google rankings. Once Google took the rankings away anyway, I figured no time like the present.


Now, in that same spirit, I've made a LOT of changes post-Penguin - deleting pages, re-organizing content, etc. I knew Google might react badly to all that, but these were changes I felt would please users, and going by the social media response to the site, I was right. So it's possible these changes caused the August 23 even-further-plummet. But not the initial Penguin drop.

On a side note, I'm pretty philosophical about this whole thing. I'm really curious why this has happened, but it doesn't really matter to me. For a while now, I've been operating on the assumption that my Google traffic would suddenly disappear one day, and positioning my sites to survive that. This experience has allowed me to test my theories about getting past a big Google drop. Now I know precisely what happens, and that I can survive it if I have to. I'd love the Google traffic to come back, of course, but que sera sera.

[Bewenched]

I've been reading a lot here on webmaster world and truly examining why our 15 year old ecommerce site could have produced a potential false positive. We don't buy links, have done some in-house SEO but mainly changes over the last year or so have been to benefit customers. The only thing we have not changed is our left hand navigation.

I've really looked into what our competitors have done with site design/layout etc and the only big thing that stands out is that we have some left hand navigation. Could it possibly be that having this site wide left hand navigation is what is producing some sort of penalty or giving us a "low quality" score of some kind?

I'm just curious if others here also have left hand navigation.

We're seriously considering getting rid of it, but I think that it may make site navigation harder for customers.

Any input on this would be most appreciated.