Why Haven't Sites Come Back from Panda? Matt Cutts Tries to Explain

[walkman]

This is a rush(?) transcript from Dany Sullivan's blog so probably not everything is 100% correct. The italics and bolding are mine.
[searchengineland.com...]

DS: Talking about Panda, says that he’s getting a ton of emails from people who say that scraper sites are now outranking them after Panda.

MC: A guy on my team working on that issue. A change has been approved that should help with that issue. We’re continuing to iterate on Panda. The algorithm change originated in search quality, not the web spam team.
....
DS: Has it changed enough that some people have recovered? Or is it too soon?

MC: The general rule is to push stuff out and then find additional signals to help differentiate on the spectrum. We haven’t done any pushes that would directly pull things back. We have recomputed data that might have impacted some sites. There’s one change that might affect sites and pull things back.

DS: You guys made this post with 22 questions, but it sounds like you’re saying even if you’ve done that, it wouldn’t have helped yet?

MC: It could help as we recompute data. Matt goes on to say that Panda 2.2 has been approved but hasn’t rolled out yet.

DS: Reads an audience question – is site usability being considered as more of a factor?

MC: Panda isn’t directly targeted at usability, but it’s a key part of making a site that people like. Pay attention to it because it’s a good practice, not because Google says so.

Matt mentions 'pull back' but that's nonsense and very disingenuous of him. Pull back to me means letting a previously labeled bad content rank. We're talking about improved sites and content, no need to pull back, just reanalyze it.

So it's clear to me that this is a penalty. Maybe if you got links from every newspaper in the Northern Hemisphere you might escape but for the rest it looks like it depends on Google engineers. It took them 3+ months to admit it.

[freejung]

The challenge we face is being limited to anecdotal reports

That's a good point -- for every report we have here of rankings not changing, there could be a dozen other sites that made changes and saw an immediate impact.

Too much speculation, not enough data...

[londrum]

Then the machine-learning program ranges over the data signals that Google collects to learn how those factors could be applied to create a more optimal SERP. When the new machine-proposed algo passes some quality checks, it becomes live.

i wonder what those quality checks involve?

presumably its a load of google employees typing in searches and visiting all the pages that come up. for that to work you'd need employees who actually know a lot about the subject that they're searching for, otherwise they're not really going to know if the page answers their question.

here's a thought... assuming that google know that this is less than ideal, do you think that's what the new 'like' button is really all about? because the people who click the button WILL know whether the page is any good.

its a tacit admission that their own employees cant quality check every search term.

[PPC_Chris]

Here is my take from the whole situation...

I think that Panda penalizes sites (or portions of sites) that hit a certain threshold of what Google considers low quality signals (thin content, too many ads, possibly scraped content, etc.).

Panda 1.0 penalized a lot of sites. Panda 2.0 and 2.1 expanded the threshold and more sites were hit with the penalty. Sites have not recovered from the Panda penalty. Panda is not running on the fly, so even if webmasters have taken action to fix the site, the penalty will not be lifted immediately but only if/when Google decides to re-calculate the penalized sites.

I believe the penalty is a %, in that Google discounts your site (or a portion of your site) by a %. In other words, all your pages now have x% of the value that they would have w/o the penalty. This differs from a -50 or -20 penalty. This explains how sites can still have the top few listings for brand-specific terms while losing rankings for more competitive terms. I have not heard of a drop in indexed pages or a sitewide de-listing due to Panda. Low quality content on certain pages of your site can cause your site to hit the Panda penalty threshold, and the rest of your site can get penalized, dropping rankings for pages that have no issues with low-quality/thin content.

I don't see how Panda can't be seen as a penalty. If it was just an algorithm change, there would be webmasters would be reporting minor loss of rankings. The rankings change would look like a bell curve - with almost all websites being affected in at least minor way. Instead, it seems pretty obvious that your site either got smacked down by Panda or didn't. Plus, many sites actually improved rankings after 1.0 only to get decimated by 2.0 or 2.1.


What I am hoping for is that at some point (SOON!) Google will re-run this algorithm and re-calculate sites that have hit this threshold and penalties will be lifted for those of us who have correctly identified and fixed the issues Google saw with our sites.

[mrguy]

here's a thought... assuming that google know that this is less than ideal, do you think that's what the new 'like' button is really all about? because the people who click the button WILL know whether the page is any good.

Or, it's a bunch of SEOs and their paid armies clicking those links. If that +1 button has a large weight on the results, then things are going to get a lot worse with Google search results.

[Reno]

then things are going to get a lot worse with Google search results.

And things are going to get a lot worse with privacy.

We know they have a record of our search queries;

We know that Google keeps an historical record of our surfing history;

They know what YouTube videos we find of interest;

We know they read emails if you have a gmail account, or if you write to someone with a gmail account, and they keep a record of the keywords embedded within your message;

If you use Google maps, they know the kind of businesses you want to find;

We know that if you have websites, they have every word on those sites identified as your ownership;

And now we know that if you give a page a +1, that you did not end up on that page by accident. In fact, you have directly indicated your interest in that page, so that is a clear signal for your profile.

Google knows more about you than many of your best friends ~ let's hope they +1 like you.

.....................

[plondon]

QUOTE: "The general consensus is that the types of links you are building have seen reduced value, and will continue to see declining value from google as it moves forward.

Also, since these kinds of links are relatively easy to obtain, it is VERY likely that your competitors are all doing the same type of link building."

Thanks Planet13.
The thing is, the markets I am in are low competition and fairly low search. And the competition isn't high quality content or high quality sites. And there really aren't that many (if any) high quality authority sites in this market to get links from.

The quality of my content is easily better than 90% of the competition, if not more. So I hear what you are saying, but even my worst links are as good as my competitors.

With that said, I was considering putting together some vids and trying YouTube and video site backlinks. Although none of my sites really lend themselves to video, so it would be a dull vid done for linkjuice only.

As a further point. While Yahoo recognises that each of my sites has 100s of links, Google still only recognises 30-50 links for each of my sites. And those 30-50 are the first links I created for each site, and are generally of higher quality.

Any ideas would be gratefully accepted, I am currently going through my onsite SEO factors with a fine-tooth comb.

(PS how do I do a quote on this site?)

[HuskyPup]

(PS how do I do a quote on this site?)

Take out the spaces:

[ quote ] [ / quote ]

[Leosghost]

To quote you put [ *quote] around the text [*/quote] without the *s ..if you click on the "preview" button , you'll find all the buttons to do the board code there .

[ThatsBoBo]

I love Panda.

All the spam sites are now out of my way in the serps for some major major keywords and phrases. >$3 clicks are typical. Life is good.

[zeus]

walkman - it can even get worse then firering people, I told my girlfriend today, it can be that we have to move out of our new House in a few Month, we just moved in in okt. but i have now lost 70% special english speaking, which is even worse for me, that was hard for me, never was hit really by a update, was gone for a few weeks then back, but this time im down from april 28 big time and i have no clue what to change, the dumb thing there was a time when i believed in not making a site for search engines, but that has changed in the last years, now you are not allowed to create a site which is for the users, which could mean duplicated contents,.....

[tedster]

It seems to me that the fuzziness in Google's response comes from the fact that they don't build algorithms the way we webmasters think they do - and they haven't for a long time, although up until now, our mental model was "good enough".

Matt may understand what "data" he is talking about re-running - but we don't. I assume that it's the actual algorithm building data itself, but I'm not inside Google.

Here's the question I have. Does Panda assign a "document classification score" when "it runs"? If so, is this something that cannot be changed dynamically, even by deep changes to the website content and structure?

And what was that earlier comment about internal link structures in Polish websites supposed to communicate? In other words, how does internal link structure affect the way Panda scores a website?

[serenoo]

That's a good point -- for every report we have here of rankings not changing, there could be a dozen other sites that made changes and saw an immediate impact.

Too much speculation, not enough data...

one week ago I changed template. Removed about 20 thin pages. Removed my directory and grouping the outbound links in only one link page. Removed the h2 text on all pages (so all pages now do not have a sub-title row). I validated many pages. I removed google adsense from all English pages. I changed some text.
Results? Nothing happened: I did not lose visitors and I did not gain visitors. I am invisible to Google.

[fakedsysadmin]

I assure you I didn't have 10,000 pages or even close, but even now I have 4 times less than I had on Panda day.

It sounds like spam to me.

[zeus]

Just made a search, got 148,000,000 results on spot 6 on page one there is a site with about 40-50 link exchanges, 20 images which all goes to a external affiliate link, so no content what so ever, they even have on the right side about 60 different Categories links which would normally go to other internal page, which it also does but with the same images as affiliate link. damn there is 0 content on frontpage , 40 link exchanges and maybe 60 duplicated content pages

[Atomic]

now you are not allowed to create a site which is for the users, which could mean duplicated contents

My experience says otherwise.

Does my site have duplicate content: Yes, most pages have duplicate content

Does this duplicate copntent rank in the SERPS: Yes, very will in some cases

Was this site hit by Panda: No, far from it.

More pages than ever are indexed. Traffic is way up. Earnings are at record levels.

[ken_b]

It seems to me that somewhwre since Panda 1 it's been stated by a Googler that they have some 500 or so tweaks coming this year.

If that's true, how in the world can anyone say any new movement in the serps is related to Panda and not some non-Panda tweak?

Maybe time to accept what is, hope for the best in the future and get on with life on the net as we know it today.

[DirigoDev]

I'm staying focused on the user - going to make a pretty little "Apple like box" and wow people with my tantalizingly perfect Web site. Funny thing is, it was close to perfect pre-Panda. Have made more change in the past 60 days than the past 4 years. All my chips are on the table. Just waiting for the cards to be drawn to see if I hit the big one or stay Pandaized.

G can kiss my ass. As I.T. Director I've preset my entire enterprise to Bing.com. My users are all good with using Bing. The SERPs are at least as good as G. I figure my 65 heavy users rack up $3/each per week in PPC clicks. That's $10,140/yr. If my users actually begin to like Bing and change their home preferences, it will cost G 50%+ more.

If M$ is going to treat us more fairly, then it it time to move our business - wholesale!

[Sgt_Kickaxe]

Google: Hey, we made a change and we're not telling you what it is.
Webmasters: Hey, I'm making changes since you made changes but it's not working.
Google users: Hey, I don't like the results anymore.
Google: Hey, we made changes to the changes and we're not telling you what those are either.
Webmasters: Hey, now I'm so confused.
Google users: Hey, I don't like the results anymore and sites I used to like are all different.

The problem can be solved if you focus down to the issue which is
a)Google not fully disclosing and detailing their changes and/or b)Google having a monopoly on traffic control which forces webmasters to obey them.

Sooner or later someone will solve one of those problems on Google's behalf, I hope they do it themselves.

[edited by: Sgt_Kickaxe at 11:48 pm (utc) on Jun 9, 2011]

[brinked]

Here is my gripe with panda so far. There are 2 sites (almost exact same site, same owner) which are very very thin affiliate sites. They used to rank top 10 about 2 years ago for a main term. They eventually were knocked down and out of sight as google updated there algo to diminish such thin affiliate sites. When panda was released, these 2 sites are now ranking top ten again. The only difference this time is that the owner removed the affiliate links from them. He used a redirect and now they are just dead redirects.

Google has rewarded these sites because it feels they are now high quality because they removed the ads. These 2 sites have 0 content at all. I mean literally they have nothing, they each have 2-3 paragraphs of completely spammy useless text.

This does go to show however, that panda has not only bumped back low quality sites, it has also given a big boost to sites it feels is high quality as well as these 2 sites were not the product of other sites getting knocked down because they were on page 20+ before panda.

[netmeg]

It may well be a temporary boost.

[HuskyPup]

G can kiss my ass. As I.T. Director I've preset my entire enterprise to Bing.com. My users are all good with using Bing. The SERPs are at least as good as G. I figure my 65 heavy users rack up $3/each per week in PPC clicks. That's $10,140/yr. If my users actually begin to like Bing and change their home preferences, it will cost G 50%+ more.

Can someone confirm/explain this to me since I think I know what has been written but just need a confirmation?

[Swanson]

"Quality is in the eye of the beholder"

That is a great statement that you need to remember - it's only quality if Google thinks it's quality!

Time will tell, however I don't believe that the ads thing makes too much difference.

I say that as I have sites with ridiculously bad ad placements (adsense) with over 50k visits per day, also with thin content.

I think there needs to be other factors that affect the situation for the ads to be an issue.

I do think pagerank and authority does help as all my sites that are fine are PR6+

[Leosghost]

Yes it indeed means what you think it does..countdown to .."the email" 3...2...

[bluntforce]

And what was that earlier comment about internal link structures in Polish websites supposed to communicate? In other words, how does internal link structure affect the way Panda scores a website?

As I recall, Matt Cutts was explaining why Panda was only implemented on English based sites. My interpretation remains that Panda considers link structure, at least in part.

I've been watching one site that Panda hit pretty significantly. No ads, clean design, it appears to have a little over 4,000 inbound links, 100 of which are random sites, then 4,000 links from individual pages on one other topically similar site.

The subject site has about 20 obvious pages, half of which lead to product searches on the site. Search pages consist of content like:
Widgets in Alabama
Widgets in Alaska
Widgets in Arizona
which then lead to pages with content like:
Widgets in County 1
Widgets in County 2
Widgets in County 3

There are 50 states and a little over 3,000 U.S. counties so the site has more than 3,000 pages with very shallow content consisting primarily of keywords and regions. The ten or so informational pages I saw reminded me of EHow and contained factual errors.

Just one site so it's just an observation.

[supercyberbob]

In the interview that the first post in this thread links to, MC acknowledges the scraper issue that is apparently being worked on.

That sounds like it wasn't deliberate, or machine learning, or testing, but a giant bug with the algo.

Also, "But there’s this mass perception outside the search industry that there’s too much low quality stuff polluting the search results. Panda was designed to address that."

Here's how I put it all together, feel free to do so with your own spin.

"We've launched a super algo update called Panda to stop the bad press we've been getting about search quality, but scrapers will outrank you in the serps."

I don't really see anything in the interview that explains why sites haven't come back from Panda after making changes either.

It seems like it took a while for the big G to admit there is an issue with the scrapers in the serps.

Reality check: Google is in it to make money, not feed the world. Does that make them good or evil?

Only YOU can decide.

[rlange]

supercyberbob wrote:
Here's how I put it all together, feel free to do so with your own spin.

"We've launched a super algo update called Panda to stop the bad press we've been getting about search quality, but scrapers will outrank you in the serps."

You make it sound like he's saying, "Deal with it." I read it as, "We've created and released an algorithm to address the quality issues that people have been complaining about. We're aware that this algorithm is not working as intended in some cases and are working to correct that issue."

--
Ryan

[supercyberbob]

rlange

hehe. Didn't mean to make it sound like "Deal with it."

My main point is it took a Q&A at SMX Advanced months after Panda was pushed out to even acknowledge the issue.

Just seems a bit sketchy to me considering the impact of the panda update.

And then there's the timeline to recovery, but that's another story.

[maximillianos]

Let's hope we see the scraper issue resolved. That is the biggest issue I have had with this update. Nevermind the hit my sites took. I can't stand the injustice of watching the scrapers rank #1.

[Freedom]

But the message is clear: have dozens of sites, don't trust what Google says and don't don't depend on them, even though they have 65%-70% of the market.

That was my reaction back in Feb and what I've been working on. Dozens of new sites in an ambitious plan to get bullet proof. Diversified ad revenue, diversified SEO techniques - all clean, but working different angles and over-lapping. Content is V E R Y L O N G.

If I have to go to 50 new sites, earning $2 a day, that's what I'll do.

It could take one to two years to complete, but spread yourself way out.

And I'm moving way beyond adsense - out of spite, and because I don't trust any googlesoft #*&! *&!0$%.

I hope the Wall street slams the crap out of them.

[SEOPTI]

They fix site quality in their algo but at the same time they give a f*** about scam artists stealing content.

Now you know everything you shold know about their priorities. It's time Bing replaces them.

Did I read in the interview with MC they dedicated a single soul to stop stealing content or am I blind? If this is the case this is just ridiculous. What the f*** are they trying to accomplish? It's really time thie monopoly is replaced by a different less evil monopoly.

I think they should completely stop communication with webmasters, they spread myths and nonsense all the time. It's clear their departments don't talk with each other. probably their internal structures are not organized and Goog Inc. will not exist in a few years. This reminds me of the Maya. And you know what happend to the Maya.

[edited by: SEOPTI at 4:41 am (utc) on Jun 10, 2011]