Instant Panda Recovery by moving pages to a different site

[Content_ed]

I'd heard quite a bit about getting out from under Panda just by moving sub-domains or pages so I finally gave it a try. I moved a half dozen pages that were drawing a few hundred visitors a day from Google on my Pandalized (down 80%) site to my Panda pleased (up over 300%) site this weekend.

It took a little over 24 hours for Google to start indexing the pages on the new site so I'm not sure if Monday results represent a full day. Of the half dozen pages, three were slightly above their pre-Panda level (year-over-year) on Monday, and three were around 20% under. The average Google traffic for the six pages Monday was around 250 visitors each.

So now the waiting begins for the next Panda cycle. I'm expeciting one of three things to happen, so it will probably be the fourth.

First: Panda could say, "I recognize these pages" and crush the entire site.

Second: Panda could smirk and say, "Took him nine months to figure that out, may as well have made a baby," and do nothing.

Third: Panda could look in awe at this great new content and triple its visitor count, like it did for everything else on the site a couple months ago.

Anybody want to fill in the fourth possibility for me?

[Panthro]

haha! subscribed for #4

[tedster]

A lot may depend on the number of pages that each site contains. If you moved half a dozen pages to a domain that contains hundreds or thousands of other pages, you may see no changes with future Panda iterations.

There's also a chance, since Panda has a site-wide influence, that these pages were not the source of the Panda problem on their original domain. In which case, you made an excellent move.

Panda could say, "I recognize these pages" and crush the entire site.

I don't think Panda uses a historical record or "memory". I think the pages will be newly evaluated as they are, and as part of the site where they now live.

[santapaws]

doesnt this show how random this panda algo is? same content is fine if the site doesnt have a bunch of other pages it doesnt like?
wheres the algo for positive discrimination for good pages?

[Content_ed]

I should have put more detail into the post, let me clear up a few points.

First, the traffic gain from Google for these pages was around 5X, quintupled.

The pages were moved from a site with around 150 pages total to a site with around 30 pages total.

I didn't mean Panda would literally recognize the content on the pages, I meant Panda might associate them with the old domain through the 301's. I don't know whether Google keeps a record of 301's or uses them and forgets them. And I have to leave them on the old site in any case for the incoming visitors not using search.

And while you may argue this is subjective, there are no "problem" pages on any of my sites. No syndication, no community content, no product reviews, other than the infringements, it's all unique, it's all professional.

But I do have a crow-eating theory I've been looking at for a few weeks that might explain Panda for my two Pandalized sites. I think everybody can agree by this time that there's no one-size-fits-all factor, they rolled out several major changes in Pandas.

I'd always believed that multiple subjects on a site is good or at least neutral. I always distrusted single subject sites myself as SEO garbage and expected search engines to do the same. WikiPedia, Yahoo Answers, Amazon, some of the hottest sites on the web are clearly multi-subject.

But here's a bit I never considered until talking it over with a friend recently. Very large multi-subject sites will tend to have an even distribution of pages on different subjects and of incoming links to those pages. It's not like all the links to Amazon focus on Harry Potter books or all the links to WikiPedia are for spinal curvature. It wouldn't surprise me if they obey the standard Pareto distribution, 80% of the incoming links and traffic going to 20% of the pages. And maybe, just maybe, Panda has tried applying a rule like this to determine website quality. It would certainly favor the big brands.

My worst Pandalized site, the one I just moved the pages off, is flipped very badly the other way. Without importing all of the pages into Word to count words, I would estimate that 5% of the pages on the site accounted for more than 95% of the incoming links and more than 95% of the traffic. WebMaster Tools keywords report includes none of the keywords from the popular content in the top ten for the site. And the subjects are entirely unrelated, about as close as Civil War Reenactment to is to nuclear power generation.

Before Panda, not only didn't it matter, it aided in some accidental discovery on the part of visitors who weren't looking for the other subjects but were engaged once they found them. After Panda? Maybe this is our greatest sin.

[dibbern2]

Great post! I am dealing with a similiar site, not apples +oranges, but more like apples + tugboat engines. No relation between the two, been a true winner for 8 years, and now 46% depleted by Panda.

[Andem]

I actually did an experiment of my own. I moved some of the most important pages of my site to a subdomain. Traffic literally tripled for 2 weeks (almost back to pre-Panda) and then dropped back down to Panda numbers. I suspect the 301s were responsible for the drop, because during the 'honeymoon' period, Google was listing both versions (the one on the new subdomain always first).

I have considered several times to move parts of my site to new domains partially because of my subdomain experience but also out of desperation. I still can't get myself to do that because over the years, I have met people who know my brand and web site address anywhere from new friends across the ocean to people I've met in airport lounges to a member of parliament.

An example worth mentioning in this thread is an ongoing experiment I've been conducting which involves a completely restructured version of a very popular section of my big site. Every single page on the experiment site is a copy of the original with a very small difference: At least one phrase or word has been rewritten (be it favourite -> favorite or 'f00bars can be found at' -> 'foo bars are available from', 'we think you' -> 'we believe you should'). Some search terms rank the experiment of scraped content in the top 10, some below the original, but most are far above what was 'stolen' from the original page. The whois info on the fake is also an alias registered at my brother's address at an address in a different country. Just like the fake's whois info, the scraper site is also hosted in a different country.

Almost like clockwork, each Panda iteration sees my experiment scraper site either completely drop or gain a lot in Google SERPS. Meanwhile, the original site with original content published between 1997 - 2011 with tons of references from government sites, national newspapers, and a huge array of PR5-PR8 sites continues to slowly decline.

Am I really forced to create a new site just because some ridiculous 'algorithm' Google 'geniuses' have developed think I'm not OK?

[atlrus]

I'd heard quite a bit about getting out from under Panda just by moving sub-domains or pages so I finally gave it a try. I moved a half dozen pages that were drawing a few hundred visitors a day from Google on my Pandalized (down 80%) site to my Panda pleased (up over 300%) site this weekend.

It took a little over 24 hours for Google to start indexing the pages on the new site so I'm not sure if Monday results represent a full day. Of the half dozen pages, three were slightly above their pre-Panda level (year-over-year) on Monday, and three were around 20% under. The average Google traffic for the six pages Monday was around 250 visitors each.

So now the waiting begins for the next Panda cycle. I'm expeciting one of three things to happen, so it will probably be the fourth.

First: Panda could say, "I recognize these pages" and crush the entire site.

Second: Panda could smirk and say, "Took him nine months to figure that out, may as well have made a baby," and do nothing.

Third: Panda could look in awe at this great new content and triple its visitor count, like it did for everything else on the site a couple months ago.

Anybody want to fill in the fourth possibility for me?

If I understood correctly, you moved your old content to a different non-penalized domain, right?

If this is true - I simply don't see how this could be a "Panda recovery", since your original website remains penalized.

[explorador]

If I understood correctly, you moved your old content to a different non-penalized domain, right?

That's what I didn't get, if you moved the pages to a new server or diff domain. Interesting try, let's see what happens.

[Content_ed]

@Andem,

I'll be keeping an eye out for the two week drop.

Do you know if the two week drop coincided with a known Panda update, or just happened on a regular day?

Google seems to have forgotten the old pages as soon as it settled on the 301s to the new pages, though I haven't gone plumbing for them with exact quotes. I'll keep an eye on that.

But I did just noticed the main moved page is back to #1 in Google for a very competitive two word phrase. It never could have ranked there based on just content, so Google is clearly applying the old link juice.

And I forgot to mention that the pages moved from a fifteen year old site to a three year old site.

Also, I did a bunch of searches on Google Japan, which seems to run current data minus Panda, and the results are pretty damn close for the pages I moved.

[santapaws]

yes they move the main content and leave the possible thin pages. And even though the old site is 301'd, it gets the pre-panda ranking of the pandalized site. defying all logic.

[Content_ed]

That's what I didn't get, if you moved the pages to a new server or diff domain

Same server, same WebMaster account.

I consider it a Panda recovery (for now:-) because the pages I chose to move represented at least 80% of the traffic the old site used to get. I don't have any particular reason to care which one of my sites the pages live on, but you could certainly argue that they fit better into the theme of the new site than the theme of the old site.

If the old site was mainly civil war reenactment and the pages moved were nuclear power generation, the new site comes pretty close to coal power generation.

[BaseballGuy]

I hope this isn't an off topic question....

But I had an "advice" website (that one one page that was ranking.....it had 12 affiliate links on it) and almost 2k words (on the pandalized page). It was hard to get unique content, but everything on that page is unique.

Do you think if I cut down the amount of affiliate links....that will "de pandilize" that webpage?

It was for many years an authoritative webpage (believe me, I was on top of my game for that page) until Panda. I have read that Panda dings sites with more than a few affiliate links on one page.

I'm just looking to get de-pandilized for that one page. The rest of the site does fine in traffic levels.

[wheel]

Do you think if I cut down the amount of affiliate links....that will "de pandilize" that webpage?

In my limited understansing, panda is site-wide. So the problem may be that page, but it could be any (or all) of the other pages on the site affecting that page, even if that page isn't specifically pandacized.

[YieldBuild]

I've seen a number of pages moved to a subdomain do well after getting hit by panda and then come down. Most of these examples, the content was simply 301 to a new subdomain.

The best opportunity to do well is to improve the content and then move it from my experience.

[Content_ed]

@Andem,

I just checked all of the pages with exact phrases and checked the additional results as well. They are no longer indexed on the original domain. Of course, I don't know whether or not they are still pulling in the pre-Panda traffic, but we'll find out tommorow if that's significant.

I also checked Webmaster Tools on the old site for the number of different domains linking to the pages I moved, it was on the order of a thousand. There were only a couple hundred distinct domains linking all existing pages on the new site. I'll try to remember to check WT every day, maybe when the pages move between them that will mean something to Google.

Has that real time Analytics thing they announced last week rolled out for anybody yet? I could use it now:-)

[BaseballGuy]

@wheel

I have heard that it's sitewide as well....Ok, I guess I need to get out my magnifying glass and have a closer look.

:)

[Planet13]

...but you could certainly argue that they fit better into the theme of the new site than the theme of the old site.

If we are going to go barking up trees, I get the feeling this is a pretty good tree to be barking up.

[Donna]

Do you think if I cut down the amount of affiliate links....that will "de pandilize" that webpage?

So far from my findings Affiliate links does not hurt , and it seems to be site wide penalty on thin/dupe content and or positioning of the content and amount of it . Now I have no definition of what thin or dupe content and how much of it counts negative .

Now I will share my case study on one of my affiliated websites Oct 14-15 Pandalization (before that ranking well):

Oct 14th total links indexed 3500, All ranks dropped if not at 950+. Some snippets could be found in SERPs some couldn't including same keywords.(Possibly discards OOP). All pages other then main page got N/A PR if it maters.


Oct 15-17 noindex all review pages and deindexed my forum a week later , removed H1 tags and xml maps. That's all .

Oct 25-27 Major UI overhaul All my affiliate links are above the fold (discounts the ads claim at least for me )!


Dec 3rd pages indexed 700 only , same day site comes back to same pre Oct 14 rankings. All pages back to Regular PR.


Dec 5th under 600 indexed.( Will keep you up to date if anything improves the more I lower the indexed pages.

Been checking longs for manual reviews , could not find any funny business IPs from google.


One more interesting thing I noticed which might help you out pinpoint issues : Webmasters Tools : Keywords : check top 3 keywords you have listed : Around Oct 14-16 my top 3 stopped including my main page , Dec 3-5 main page was again on top for each of the keywords.

First I thought its another OOP but then I found out the panda tweet and all the commotion. Prior to that date no aggressive link building or any onsite optimization. Not dancing for sure.


My conclusions :

If it was a panda hit which I think it was reduce indexed or useless content , don't kill your ads as that will change eventually your overall website profile in google's eyes. Zoning is actually important as I stated previously in another post, noticed some funky stuff on one other website of mine experimenting currently with it. Dynamic website are having very very tough luck right now compare to static pages another issue you might look into. It took me 3 weeks to recover .So far its been 4 days with same old results pre Oct 14th.


Please share similar activities it will help many pinpoint problems . Different niches seems to be treated differently .


P.S. Never been Pandalized on any of my websites prior to the 14th of Oct.
P.S.S If I open the floodgates I could easily have more than 10k pages indexed but at what cost.

[tangor]

Sorry, kiddies. Looks like smoke and mirrors to me. And possible duplicate content (which hits the originating site more than the "newbie pages". Reality is that older/elder sites are going on the back shelf just like Ivanhoe or Moby Dick... UNLESS there is absolutely FRESH, NEW content on a regular basis. What benefit of a two/three week boost with no future? With 301 redirects that track back to the original site? Can't see that as a way forward. YMMV

[Content_ed]

@tangor

Are there any documented cases of other people doing this? I didn't research carefully, all I saw was people playing with sub domains or moving a whole site. To me, there are obvious reasons neither of those approaches would make sense.

I don't have high hopes for this being a permanent fix, but if it fails, it will be because Panda is keeping track of penalties assigned to a site on a URL-by-URL basis, not because any algorthimic "quality" issues with pages in their new location.

I've also noted that over the last two days that the pre-exisiting pages on the site have seen a solid traffic boost from Google, the previous highest ranking page is up nearly 40% week-on-week. Unless I missed the announcement of another Panda update, the only reason I could assign is that the authority of the site has shot up with the new links.

Look, I'm not in a position to argue about what Google thinks is smart vs what they think is suspicious. It's clear to me that Google and I don't see the world the same way. I do know that the pages I moved were basically the top destination in the world for their subject matter, pre-Panda, with great organic linking, media citations - tens of thousands of copyright infringements can't be wrong.

From technical standpoint, based on what Cutts had to say about how Google treats 301's, there's nothing wrong with moving pages from site A to site B, and given the better theme match, I think it's the sort of thing Cutts himself would have suggested to me a couple years ago if he happened to know me and the sites.

In any case, how anybody can predict a two/three week boost based on two days of data is beyond me. It could crash tommorow, it could crash next February on Panda's first anniversary, or it could grow for the next fifteen years, like the old website. My idea here was to share real data, something I wish there was more of on the board.

[ScubaAddict]

I'll add a bit here. I broke my site into subdomains based on about 6 major subjects (301'd all content to new locations). I figured this could help me separating good from bad content and let the bad content (in google's eyes) sink itself and the good content would rise.

I also saw an almost immediate recovery to pre-panda traffic for about 1-2 weeks for all subdomains. Then the content that I thought was super-excellent dropped like a rock (down 70-80%) and the content I thought could be considered marginal dropped by only 10-30%. Overall traffic evens out around 60-70% down - my pre-subdomain levels.

Long/Short - Subdomain to re-pandalization was about 2 weeks (as everyone else states). Doubt a new domain would be any different - but would be interesting to see.

[santapaws]

but there are sveral comments on new domains working long term and subs not working long term. This isnt new.

[ScubaAddict]

@santapaws, I appologize - there are actually reports of new domains working to recover permanently from Panda? I haven't been reading much since I have given up on my site...

[dunivan]

I think you will end up getting the new content pendalized and will see a decrease of the new content.

If that happens, I am wondering what adding a rel=canonical tag will do to the content in its original place.

[SEOPTI]

A site-wide 301 in most cases results in a -50 "goog full of paranoia" penalty. At least it raises a flag and the website will be inspected by a manual Goog owned and underpaid Indian or whatever pest inspector.

Don't do this with two domains at the same time (footprint), they will kick your ass (I mean the pest inspector will kick your ass) if you try this, you wil find yourself in the -50 hell with all your domains in your webmaster tools account. Warning! Even domains which did no 301 redirect will be send to -50 hell!

They need to do this because a 301 will NEVER transfer the original penalty. (reason: dumb algo which means dumb people, dumb goog engineers)

Believe me, this is 100% experience based, no conspiracy theory, this is fact!

How many slave spies from Goog are reading this? I believe at least a dozen. They all want to earn more money and will report back to the Goog Gods.

Solution: Bing it! www.bing.com

[dunivan]

At least it raises a flag and the website will be inspected by a manual Goog owned and underpaid Indian or whatever pest inspector.

Manual graders cannot influence ranking. Also, a site wide 301 redirect is the recommended fix for a changed domain.

if you try this, you wil find yourself in the -50 hell with all your domains in your webmaster tools account

the thought of this is crazy, google will not cross penalize just because a site is in an individuals WMT, it is too arbitrary.

[nadavs]

@dunivan - unfortunately, SEOPTI is right. I changed my domain because I just wanted to change the name (Google brought great traffic and I was never hit by any Panda). Three days later traffic started to drop, and three months later I'm still struggling.

My site is still in the fifth page for most of its keywords, only hoping to rise some day.

nadavs

[Andem]

@Content_ed: The two week drop was not exactly on a day that was reported on webmasterworld as a minor Panda update. If I remember correctly, it was within 1-2 days of a reported Panda update, though.

Has that real time Analytics thing they announced last week rolled out for anybody yet?

I've removed analytics code from all properties for a list of privacy reasons. Running your own stats is pretty hefty sometimes, but worth it for the instant information you can see and aren't forced to share. I'd suggest running webalizer or something if you want results right now.

@ScubaAddict: EXACTLY my experience as I mentioned. It would be nice to see how Content_ed's experience works out though with a brand new domain which is more focused. I'm definitely stayed tuned to this thread.

[dunivan]

@dunivan - unfortunately, SEOPTI is right. I changed my domain because I just wanted to change the name (Google brought great traffic and I was never hit by any Panda). Three days later traffic started to drop, and three months later I'm still struggling.

My site is still in the fifth page for most of its keywords, only hoping to rise some day.

nadavs

because you moved your site or because of a "site affiliation" in WMT?