New Panda Update - September 27

[BrodyDodes]

Google rolled out its most recent iteration of the Panda algorithm on Wednesday 9/28/11 (or Tuesday according to who you ask). Saw the first recovery of one of my punished sites since the first iteration back in February. Anyone else see recoveries?


[webpronews.com...]

When asked if an iteration of Panda was implemented this week, a Google spokesperson told us, “yes.” She also provided the following statement:

“We’re continuing to iterate on our Panda algorithm as part of our commitment to returning high-quality sites to Google users. This most recent update is one of the roughly 500 changes we make to our ranking algorithms each year.”

If you’ve followed the Google Panda update saga throughout the year, you may recall Dani Horowitz’s story. She runs an IT discussion community called Daniweb, and it was hit hard by the Panda update, but she made a lot of changes, and gradually started to build back some Google cred

[digic]

@whitney and SocietyRoyalle, thanks for that.

many of them contained only a single sentence

This is pretty related to my site. Since I was hit by Panda mid of June, I immediately removed useless pages with less than 100 words. Also did some adjustment, but 'til now didn't see any recovery.

I also have those pages with a single sentence for a question (a wordpress site turned into Q&A) but with around 500 to 1k accumulated words on comments. All in one page. Does G sees it as thin content? Are comment/replies counted as content?

[Donna]

Expect another shift very soon.

[Shatner]

@Donna Why do you say that?

To me the most depressing thing is that I've been following the Panda saga since Panda 1 hit on Feb 27... and nothing new has been thought of or said since Feb 29. :(

Still after all this time no one knows anything.

[Sgt_Kickaxe]

Still after all this time no one knows anything.

We know that Eric Schmidt has a fondness for large corporate websites, even if they only have user generated content (ebay, youtube etc).

[Donna]

From my observation on the KWs I follow, its seems that on the 26-27 Panda/penalized sites were let back in and then on the 28th the hammer dropped and for 15 on my terms 40-50%(long tails) of the top 100 results were removed. The number is exactly between 40-50%.

I mentioned a while ago something big happened at Google end of July , start of Aug which I bet is the trigger to this latest update. I expect this to lead to an actual bigger update . Somehow everything is tied up to % content x pages x ads x authority and not backlinks, but backlinks at the same time are the main driving force . Have you noticed how strong EMD are these days ? In my view all the talk about quality content is really a big sack of CRAP. My most unoptimized sites rank the best ever with really poor content till my unique high quality content barely makes it.

Moral of the story : content =/= the king , backlinks build in very moderately will win the day . Don't you see it everyone from Google is just spreading propaganda to lead you the wrong way while they try and figure out the mess they created. I am 100% sure all those panda demotions are automatically detected but manually triggered. Just look at the trend it's so obvious. Google is in the weeds.
A lot of people will say yes but after Panda 1.0 I did this and that and gain some traffic back, well you did because your content became better and you attracted more links, not because Google saw your content getting more rich or interesting.

We are just reinventing the wheel all over again but the long way around . Crate a quality content and they will come (says Google) . Well build your damn links and they will come even faster (says me)but really really moderately.

[tedster]

those panda demotions are automatically detected but manually triggered

I really doubt that the Today Show or Motor Trend were manually triggered demotions.

I expect this to lead to an actual bigger update

Did you see this prediction from Brett Tabke? [webmasterworld.com...]

[Donna]

Tedster how about possibly forcing specific/unique content with the main goal of boosting OBLs for stronger ranking purposes and could that be possibly the main goal of panda and restructuring the entire way of linking around. Like you have the choice but who will you vote for and why and what will you gain from it ? How would you cheat the system ? And how is the system going to see you. Pretty much reverse link juice, you can do whatever you do but to the certain limit where limit = ranking.

[Shatner]

From my observation on the KWs I follow, its seems that on the 26-27 Panda/penalized sites were let back in and then on the 28th the hammer dropped and for 15 on my terms 40-50%(long tails) of the top 100 results were removed.

Incorrect. My site was pandalized at almost exactly noon PST on the 27th, according to all my stats, the Google traffic immediately fell 50% at that exact moment. Someone flipped the switch then.

This does not explain why you think an update is coming soon. All past data indicates that Google NEVER runs a Panda update more often than every 3 weeks. So at BEST it's 3 weeks before there is another update... and maybe more... and even if there is an update history shows that at least 50% of the updates don't seem to have any actual, significant effect.

Somehow everything is tied up to % content x pages x ads x authority and not backlinks, but backlinks at the same time are the main driving force.

Proof? I've been following this non-stop since February and I've yet to see any hard evidence to support this assertion.

Certainly not enough to justify your seemingly hard conclusions.

In my case, my site meets all of the metrics you used (and yes I've had 3rd parties confirm this, so this isn't just my unobjective view) and it hasn't made any difference. I even ran a couple of months without any ads at all, and after a couple of Panda runs left me pandalized gave up on that as well.

In my view all the talk about quality content is really a big sack of CRAP.

That's one thing which I think has been proved. Quality content is at best, one small component... and even if you have quality content you can still be pandalized.

[Donna]

Proof? I've been following this non-stop since February and I've yet to see any hard evidence to support this assertion.

My observations are only on 1 specific niche . Very competitive and very profitable.

[whatson]

My sites that have thus far survived panda have the oldest, most out of data content out of all of them. I am just too worried about updating them, in fear of what will happen. This is just an insane game to play on Google's behalf.
My sites with absolutely fresh professionally written content by an authoritative on the subject, are in Panda hell.
However, I have also noticed my non-panda'd sites traffic down perhaps about 10% or so since this latest panda update. But my major positions still appear to be in-tact so think they are still safe. Prior to the update though, the traffic did spike.

There is some seriously weird #*$! going on with google, it would be exciting and interesting if it wasn't for the fact it was our lives they are toying with, so instead it is painful, stressful and nauseating.

[chrism]

Now I'm confused, several of the sites that were hit for us at the time Panda ran dropped as you would expect with a 'victim' site.

Strangely, some of them are now recovering, without any additional work being done on them. I have made changes to two sites, neither of which have moved at all. It's probably too soon for them to have been crawled, indexed and reranked anyway, so I'm not reading too much into their lack of movement yet.

I was under the impression that after a Panda run people seemed to stay down wherever they got dropped to in the SERPS, but a significant proportion are now ranking approximately half way between their pre Panda run position and where they initially dropped to.

As a concrete example, a site which ranked #2 dropped to #15 almost in front of my eyes on Tuesday (UK), then stayed there until today. Since I got up this morning it has returned to page 1, at #6.

I'm sure someone will tell me this is normal Panda behaviour, but I thought I'd mention it as I don't recall reading about this pattern before.

[pontifex]

Just some short note in the aftermath from me:

traffic kept the same, revenue dropped by 10% in average compared to the last week, same time frame.

Panda 2.5 got me on the wrong side now: less traffic was compensated by better conversions the last 2 times.... now the traffic is solid, but the conversion dropped badly! Grrrrrrr!

Stop jumping around unpredictable, stupid little search engine!

P!

[courier]

For every looser there has to be a winner, a site I created around a year ago - keyword domain with 6 pages of content has moved to #1 for a very popular and large search volume keyword. To all aspects this was a dormant site, with no work except the initial pages for a year and nothing being sold on the site despite the commercial aspect of the search term, I never got round to arranging something to sell on it.

Is Google saying to us then, put a site up and do not update it and they will reward you. Do I begin adding content now or leave it as it is?

[gehrlekrona]

@courier, LEAVE IT as is for now!

I had a "spike" yesterday and to day seems REALLY SLOW from Google :(
So, did we have an new Panda bite yesterday?

[HuskyPup]

I'm seeing some major movements in my sector with many of the Alibaba type sites that have dominated since Panda 2.1 being shoved to the bottom of the first page plus I'm also seeing shopping sites re-entering.

I have stuff going up and down, for instance a widget term I was #1 since 2000 was demoted to the second page, then slowly came back up to #8 where it's been for a couple of months, today it is back at #1.

My biggest traffic url channel has had very little traffic today compared to usual, so far about 10% of my last month's average...we'll know soon!

Must look to see why the Panda thinks those new sites deserve to be at the top.

[Dave_Hybrid]

Im going to give no ads a full try for a month or two, got some savings to get me by. It's the only thing i have left to try.

[Shatner]

Im going to give no ads a full try for a month or two, got some savings to get me by. It's the only thing i have left to try.

I wouldn't hope for much. I tried that after I got hit by Panda back in February and it had no effect. Eventually I put most of the ads back and then I recovered (only to have that recovery randomly taken away)

I was under the impression that after a Panda run people seemed to stay down wherever they got dropped to in the SERPS

As someone who has been Pandalized 3 times now I can tell you that there's always a little fluctuation shortly after each pandalization. There's a HUGE drop, then some little adjustment... for me that little adjustment is that my traffic always sinks a little more over the next couple of weeks, but I guess it's possible that for some it could go the other way.

[nsqlg]

I have medium/large site (in Portuguese) hit by Panda.

I has before May/2011 something near 250K visitors daily, today is something near 110K, we lost 60% of all Google traffic.

Our site have more than 10 years, never got any penalization before... but early this year we migrated the domain to rebranding (long history) and this nightmare began....

In 27/Sep we have huge spike (100%) in G traffic, but back to "normal penalized" levels after that.

Our site is a mix of articles/online books written by us and forum. I decided put noindex on all poor performing pages. IMHO will be more hard to users which need find uncommon content/pages...

[honestman]

While there are many penalties clearly assessed by identified Panda versions -- and as many intelligent theories as to modes of recovery -- I wonder about one particular possibility among the many others.

Is it possible that Panda does not like sites which link out to many other sites as an editorial service to the user? In other words, I may say that we have a lot of information about a given topic, but not all possible information and angles about everything relating to that topic (impossible to do, in my view). As a result we link out to sites which we believe may provide such information (e.g. certain Wikipedia pages). Am I linking out too much? We are not a link farm at all, but we believe that linking out is part of our editorial duty and we were respected pre-Panda it appeared. Must I assume my role as an editor of what we perceive as a useful link in a field in which I am an expert is no longer seen as an appropriate role? In other words, does Google always know better than we do on every subject according their algorithm?

In addition, we now do queries which return specific information rather than general information, which makes research far more difficult, as specific information on a general search term often returns SERPS for a given brand or company and not websites which list and evaluates such brands or companies. Is it no longer possible to be a niche review site with much original content without being penalized?

[Shatner]

Is it possible that Panda does not like sites which link out to many other sites as an editorial service to the user?

I've been wondering about this too. One thing big brands have in common is that they don't usually do this, while smaller, independent sites do.

On the winners and losers list... most of the winners don't do this while most of the losers do.

[netmeg]

I don't think you can look at one thing like that, and think that's the problem (or that's the solution) You have to look at everything as a whole. Forest, not trees.

Most if not all my sites link out a *lot* and none of them has ever been hit. Personally, I think it's probably a quality signal to link out to a quality site. But it's only one signal among many.

[Shatner]

@netmeg A Forest is made up of trees. To understand the nature of the forest, you must know what kinds of trees are in it.

Panda wants a very specific kind of tree in your forest.

Obviously that one thing is not the magic fix... but it might be a component.

If all the winners have a policy of not linking to competitors, then it could mean that Panda takes into account how much link juice a site has flowing out versus flowing in.

If every article on your site links to someone else, then you have way more link juice going out than a site that does not.

[Atomic]

@Shatner - Sometimes the only way to understanding is to view the whole which makes obsessing on the parts a fruitless endeavor.

[azlinda]

Bing here I come. Yesterday my CTR went down 50% and it remains there. I've been battling AdSense since
April 11, and it is not worth it any longer.

[honestman]

@netmeg and @Atomic.

I quite agree, and I guess failed to make sufficiently clear in my intro that I was speaking of one of many possible factors. I like to look top-down where possible as well. But when I notice a pattern wherein sites remain largely self-referential and are not penalized, while many sites which link out to provide services to their users are, I have to wonder we should reduce the amount of outgoing links. I know, of course, that there always has been the famous formula of outgoing versus ingoing link juice, but that never applied to us pre-Panda due to the quality of sites linking in to us, I believe.

This is important in our case, as we will totally change the way we present our information if outgoing links are considered a strong negative signal. There are always exceptions, but I am seeing many sites not even hyperlinking (including large new media sites) to the sites they reference.

And we are talking about links to quality related sites/subjects in every respect.

Some of the links are organically reciprocal (we do not seek link exchanges), but all are meant to enhance the "user experience" in the broadest sense.

No doubt there are up to 500 other things we are doing wrong according to Panda, but we are working on them in accordance with Google's guidelines incessantly without recovery, while Bing/Yahoo have remained consistent for quite some time -- though that could change as well, of course.

By the way, I have noticed the same issue with internal linking. It seems that many of the pages where we link to related internal content are not doing as well as they once did by far. Perhaps internal cross-linking is also frowned upon.

[danijelzi]

Is it possible that Panda does not like sites which link out to many other sites as an editorial service to the user?

In my case that's true. In articles on a pandalized site I have dofollow links to other relevant sites and pages, about 1 link per article.

However, other sites in my niche also do the same and rank well so I'm not so sure if that's the problem.

I would like to suggest that everybody double check their outbound links. Over the last five years, I published about 2,000 posts on the site, and recently I did a link check and found that I had more than 300 broken links to external pages, mostly 404. When I add to that about 400 ugly amdfhxyz.com-like affiliate links accumulated over time, my outbound link profile looked pretty bad. I fixed that 404s and removed aff links, so we'll how the next Panda will react.

[tedster]

I can't see how presence or absence of external links, in itself, could be a factor. No external links? I know sites that were hit by Panda and many more that weren't. Lots of external links? Same thing. Just a few external links? Ditto.

The quality of pages that you link to "might" be an issue, and the motivation behind your internal linking as well (are they really helpful for visitors, or are they there just to push PR around?) But I personally would not look in this area for helpful fixes at all.

[honestman]

@tedster

Thank you. I will take your word for it in terms of external links. You probably have looked at more sites than I in your role.

In terms of internal links, there is/was no attempt to push around PR.

Both forms of linkage are purely designed to enhance the "user experience" and are completely "white hat" and researched to be so. Our users have commented on the usefulness.

Off to look elsewhere I guess (he says pounding his head against the wall).

[johnhh]

I would say there is a difference between linking to Wikipedia , which then links to other external sites, and linking to those external sites themselves.

For the record, we don't have links to Wikipedia, but they have links to us.

I don't see see this as a major Panda factor, there are other things at play. Google has already said, I believe, there is nothing any affected site can do. However, this is probably not true, as it's just a computer program.

[SocietyRoyalle]

Is it possible that Panda does not like sites which link out to many other sites as an editorial service to the user?

I don't know the answer to this question, but on our site which just recovered in this latest Panda update has links going out on nearly every page.

We believe it's an important service to our readers to link to further information - some links will be internal, but there's usually at least one external link.

If we publish a story on a new product we nearly always link to the manufacturer's product page - occasionally we'll also provide non-affiliate links to a retailer that sells it, and occasionally even an affiliate link to someone like amazon.com.