Hit on 2/24/2011 and come back?

[walkman]

According to Amit Singhal, Panda was a specific algo change and after that many others took place. So 'Panda 1' and 'Panda 15' were not actually Panda, but different algo changes while keeping Panda on top of everything else.

Some publishers have fixated on our prior Panda algorithm change, but Panda was just one of roughly 500 search improvements we expect to roll out to search this year. In fact, since we launched Panda, we've rolled out over a dozen additional tweaks to our ranking algorithms, and some sites have incorrectly assumed that changes in their rankings were related to Panda.

Also Matt Cutts said that Panda data is not collected daily but when they decide to run that part of the algo.

short version is that it's not data that's updated daily right now. More like when we re-run the algorithms to regen the data.

So it appears it's like a score /grade and it sticks for a while for those already ranked. How often they do it is everyone's guess but we know it can crush your site rankings.

Now we hear stories about people improving their rank here and there, but most so far have been hit in April and in the following updates, not on 2/24.

So the question is: Has anyone hit by the 2/24 algo made a comeback?A serious one I guess, close to pre-Panda traffic levels not the 10% plus or minus.

I'm trying to see if there's a set time penalty or not. If you came back from 2/24, did you by any change contact Google or posted your site in their forums (I refuse to believe that no manual changes are made to it)

Edit: My 4 sites that were not hit by Panda got an 25%-30% boost today, my one Pandalized site kinda down a bit despite many changes.

[ErnestHemingway]

This is a very tricky update but mine were more impacted during May 5th Thursday than any other update. That makes it about 5 days ago and so far I have not seen any changes.

One thing is certain if you are not making it easy for someone to know that you are real then you may be ranking high for now but as the filter runs next time then you might be the next shot.

Make it easy for someone to know that you are real person behind the site.

[Shatner]

Really good observation walkman.

It does seem like most of the people coming back were hit in the March update.

For my part I was hit the worst in Feb with Panda 1, and then it just got worse with Panda 2, and I haven't recovered at all, not even back to Panda 1 levels.

I kind of feel like Google's whipping boy, that they cut a piece off of whenever they want to give traffic to someone else.

Have any of the orig Pandalized crowd come back at all?

[Pjman]

I haven't heard of anyone gaining any greater than 10% ground if they were heavily impacted (50%+ referral drop) on 2/23 9PM EST. Outside of anybody who did not get any press coverage as being unfairly targeted.

7 of my 10 sites got Pandalized on 2/23, I just see random fluctuations. All 7 had a substantial amount of user generated, thin content. I removed or beef up all that over 2 months ago, still no substantial gain.

I think we're waiting for the Phantom of reevaluation to visit us.

[irock]

Hi, i am not sure if it's appropriate to name someone else site on this forum. But I know of one site that apparently has come back from late Feb Panda 1 update around April 15th. The site is now back to pre-panda level (I compared to their last year April data.) I have confirmed this by checking their 'quantified' quantcast data.

Moderators, pls let me know if it's okay to put website name in this forum.

[sailorjwd]

As of this afternoon I'm seeing major improvements for the 20-30 phrases I usually track. One in particular which I've never been able to get above position #40 is now #6. I'm nervous that G is just showing these results for me even though I logout of Google before trying them. I'm not seeing significant increase in visitors likely because my 1000's of long tail searches may not have improved (yet).

ps. I was hit by 35% on Panda 1 and another 15% over the next several months gradually.

[Shatner]

@sailor I don't think phrase position improvement means much. What we all really need to be tracking is overall traffic from Google.

@irock The site you're referencing, have they told you they recovered or are you just looking at Quantcast? That's not always accurate. In my case for instance my Quantcast data looks better than it is and isn't really reflecting the Google drop as you'd expect.

[Asia_Expat]

I don't think phrase position improvement means much.

Really?!... One of the most important phrases for my site hasn't got back to #1, and is still down at #9... that only brought 20 or 30 new visitors per day when it was at #1, but it was super high quality traffic, laser targetted, going to great content, and was very sticky. It's those low volume, highly important ones I need back... not a bunch of cruddy traffic to forum threads that immediately bounce away from the site.

[edited by: Asia_Expat at 2:49 am (utc) on May 10, 2011]

[dickbaker]

I haven't recovered from February. If anything, things have become worse.

Google was always 44% of my referrals, +- 1.5% or so. After 2/24, it dropped to 28% and held steady. It's now dropped to 25% in the last week or two.

I'm making changes, but the changes to each page are taking me a day or more and, with 2000+ pages to do, I suspect it's going to be a long time before I can measure any changes for the better.

[Shatner]

@Asia It doesn't mean much as a concrete sign of recovery. People have been reporting all sorts of fluctuation for invidual keyword phrases, but nothing that leads to any concrete, increase in overall traffic.

So when someone asks "Have you recovered from Panda?" saying "I recovered for a keyword!" doesn't really answer the question.

[sailorjwd]

@shatner, I agree about the long tails. But for a consulting biz I need the folks searching for specific services. The 97% other long tails are good for Adsense but only marginal benefit to business customers. Total visitors may have been up 10% yesterday.

[walkman]

The 97% other long tails are good for Adsense but only marginal benefit to business customers.

Not really, for e-commerce especially. A search like "Black Leather Sofa with three recliners" or "green shoes for brisk walking" can be a gold mine. Not everyone can rank for 'green shoes' or 'leather sofa' so long tail was a decent way to get some visitors, but now it's all or none.

[agent_x]

I wonder if the Panda algo was only run on certain sites, e.g. those in the top 20 for certain keywords?

The reason I wonder this is, I had two sites recover yesterday - but not from Panda. They were both hit back in November with what I suspect was a duplicate content penalty - duplicate content shared across other sites in the network. So I set about fixing that issue but left both sites pretty much alone, and they both really need an overhaul. I am sure if Panda had looked at them they would have been marked down, but one of them now has more traffic than it ever had from Google. Now if this one deserves all this traffic then other sites in the same network certainly do but they've been pandalised and this one hasn't. So maybe it slipped under the radar. Or perhaps it'll get killed off again soon, I'm not holding my breath.

[sailorjwd]

@walkman. right. WMT used to show 8000 keyword phrases for a month and now down to 5500.

[browsee]

Just saw the post in SearchEngineLand, Google released Panda 2.1 recently. Has anything changed?

[edited by: browsee at 8:13 pm (utc) on May 10, 2011]

[outland88]

Actually since day one I've thought your estimates were rather optimistic Walkman. It's good to be optimistic. My interpretation lately has been it may take a whole new rewrite of the page. Based upon about eight reports now even minor SEO of a title tag could send the already lowered keyword terms lower. That could vary with the traffic levels for the term and I would not think every area receives equal treatment.

Some publishers have fixated on our prior Panda algorithm change, but Panda was just one of roughly 500 search improvements we expect to roll out to search this year.

Is it any wonder that only a brand, a site with thousands of BL's, or a less than competitive area could hold up to the onslaught. Those are some odds remaining untouched.

[edited by: outland88 at 8:12 pm (utc) on May 10, 2011]

[tedster]

Here's the link to that new SearchEngineLand article (published today): It’s Panda Update 2.1, Not Panda 3.0, Google Says [searchengineland.com].

Panda 2.1: Google won't release the percentage of queries impacted but says this is far less than in the other updates. Changes were made in the past few days.

[outland88]

Yeah, I mentioned in the forums they did so. They were actually coming back for what they had missed first time around.

Panda 2.1: Google won’t release the percentage of queries impacted but says this is far less than in the other updates. Changes were made in the past few days.

Well its sorta thay way when you've shot quite a few in the room already.

[tedster]

Yeah, I mentioned in the forums they did so.

Yes, you did, along with Crobb35 - credit where credit is due for you both. Outland88 called the new Panda tweak as beginning on May 5.

There does seem to be something brewing since early May 5th. It reminds me of Adwords where I had terms ranking 10/10 and were not even vaguely related to their quality rules. Make changes to improve the quality and things tumbled and prices went up. In others words it became quite clear Google couldn't determine quality or even junk. It was more about matching their criteria of what is, for lack of better words, "good".

[webmasterworld.com...]
Danny Sullivan's article links to the SERoundtable article which then links back here. The observations you folks made here are the primary source that Google then commented on to Danny.

-----

To return to the primary topic of this thread, even with two follow-up Panda changes, it sounds like no one affected by Panda (any number) has actually recovered even most of their traffic, right?

[Shatner]

So we've had 3 Panda releases that we know about now, and neither after Panda 2.0 or Panda 2.1 did anyone recover from Panda 1.0.

So does that suggest that the algo only applies the penalty when it runs, it doesn't remove it, and that the penalty (even if it's not exactly a penalty... it acts like one in every way so let's call a spade a spade) is in some way permanent or at least locked in for a specific period of time irrelevant of whatever the Pandalized site does?

It seems to.

[walkman]

Actually since day one I've thought your estimates were rather optimistic Walkman. It's good to be optimistic. My interpretation lately has been it may take a whole new rewrite of the page. Based upon about eight reports now even minor SEO of a title tag could send the already lowered keyword terms lower. That could vary with the traffic levels for the term and I would not think every area receives equal treatment.

Well, I have done that, and deleted so many pages that it's a just a shell of what it once was. Compared to my competitors my pages are now fine, better than most in fact. I did spend lots of time doing that on the first month. Google had said, remove thin pages, improve and the algo will pick them up.

Not so, looks like they are still just piling on penalties and giving confusing 'advice.' This Panda 2 and 2.1 is totally different from Amit Singhal said for example (see his statement on my first comment,) but hey it's Google, known to not make sense or be consistent. If they ask people to do changes and then send them to oblivion for the same changes, what can I say? I thought that after picking up the changes a few crawls they'd recalculate again and re-rank. This much even JohnMu from Google said.

I assure that my competitors didn't improve that much and my pages didn't get that worse for my site to keep sinking.

Yes Shatner, it seems to me that they are only adding sites ti the 'Pandalized' bucket, not letting anyone go. I find it impossible that no one got their sites better and recent changes here have shown that people do talk when their traffic improves.

[indyank]

Has crobb305 recovered? He was one of the few exceptions who said that he got back 70% of traffic.

[indyank]

Though they say that some are fixated on panda, they do call it Panda 2.1! So they do confirm that tweaks are being made to the panda algo.

By calling it 2.1 and not 3.0, they might also be saying that a major one is being worked on.

Danny seem to be confirming that any news on a major release will be available on SELand first and we may not see anything that big in the next few months!

[edited by: indyank at 4:09 am (utc) on May 11, 2011]

[drall]

I monitor a few hundred sites in my niche Tedster and none have recovered, all are flatline since the first Panda and some have fallen even further.

I would say combined these sites lost in the tens of millions of visitors a month range.

[reblaus]

We have not recovered and since this week i have reverted the massive page deletion of "thin" pages. I do not care anymore. I just do not agree that a short article/page is bad quality. I follow Google's advice and forget that there is a Search Engine.
We have made tons of other on page changes that have been discussed here that I find have been good things to do, but deleting our history is not acceptable.
Note: we are a news site and not a site with a defined set of pages.

[walkman]

On Google support forums another reportedly came back but he posted on Google forums since day one so maybe they looked at his site manually. He was 30% down on Panda and now is back, more or less.

Maybe we haven't come back because we listened to Google post-Panda and deleted pages and then they changed their minds? Sad, very sad. No one has a clue.

Check this advice Matt Cutts gave last year and compare to why we think Panda smashed our sites: [youtube.com...]

"I believe if you were to talk to our crawl and index team, they would normally say "look, let us crawl all the content, we'll figure out what parts of the site are dupe (so which sub-tree are dupes) and we'll combine that together. ...I would really try to let Google crawl the pages & see if we can figure out the dupes on our own. "

[edited by: walkman at 4:58 am (utc) on May 11, 2011]

[supercyberbob]

Wow yeah, don't delete content because Google told you to do so.

Next thing you know, they'll have you on all fours barking like a dog, or worse.

They haven't been very transparent about the time frame on recovery, AND, nobody's regained significant traffic after following their super new guidelines.

[koan]

Putting a googlebot noindex tag on pages that you have doubts is probably a safer method than deleting them if you want to revert back in the future.

[walkman]

Do you guys think they ran it on April 11th and May 6th or just added new penalties?

I am beginning to think it's like Page Rank on the toolbar. Algorithmic but run when Google decides and at different intervals. The only difference is that there's no 'internal page rank' that updates daily in this case, you're stuck with the score from the last time. Google probably did the PR thing to stop SEOs from seeing their PR go up as they got new links. The same principle could apply here, even though many just did what Google said it's Ok to do, and got penalized.

[rlange]

tedster wrote:
To return to the primary topic of this thread, even with two follow-up Panda changes, it sounds like no one affected by Panda (any number) has actually recovered even most of their traffic, right?

My company had one site hit by Panda 1.0. This site was hit again by Panda 2.0 along with 5 other sites we own. Two of those additional sites recovered right around the time of Panda 2.1. The others saw no noticeable change.

(Edit: I made no changes to the sites that recovered and only per-page tweaks and clean-up on a handful of pages on the site hit by Panda 1.0.)

--
Ryan

[edited by: rlange at 1:42 pm (utc) on May 11, 2011]