Three sites, same publisher, different results from Panda

[Content_ed]

I've posted a few times about how two of my three simple, HTML content based sites got slammed in Panda, and one had a slight gain. Try as I might, the only differences I could find between the three were the age (the unaffected site is the youngest), the level of infringements (the youngest sites has been largely untouched), and the subject.

I was reading the WebMaster Tools thread today and saw a post by a large magazine publisher, I'm not sure if we are allowed to use names here, but they publish a number of high quality magazines for do-it-yourselfers and wannabes titled "Fine ______". According to the Webmaster Tools post, two of their sites have been impacted, the remaining eight or nine are untouched.

As near as I can tell, the platform and structure of these sites is basically identical. Same layout, same ad programs, same forums, etc. The main difference I see is with the subjects.

When Google say 11% or so of queries have been impacted by Panda, is it possible they chose a few subjects that are notorious with article farms and only applied the new filter to them? And if the results of the filter looked suspicious, penalized the whole domain, even when the domain included much more?

[tedster]

Everything I see and hear tells me that Panda is a web-wide quality detection algorithm and not something applied to only a few verticals.

In a way, you are in an enviable position, at least as far as doing analysis is concerned. I'd suggest looking WAY outside the normal SEO box for other possible factors.

[falsepositive]

Was about to post something on a different thread, but saw this and got intrigued. I too have 2 diff. sites, one survived and gained a lot of traffic, the other fell and lost. I run both sites. Ironically, the site I care about and really tend to was the one hit. It was the bigger site and I spend 90% of my time on it. It has the superior content and has much better design, etc.

The differences I see: winning site - smaller, simpler, lighter. No link profile, no links in and out, nobody cares enough about it to duplicate it in any way. It's under the radar.

Losing site - big, popular, authority, visible, strong link profile, more ads (yes, but not that much more), more monetization (understandably), more community, scraped a lot.

If I simply ogled at both with my own eyes, the differences are night and day. These sites occupy the same niche. User metrics also echo what I believe: the winning site has inferior metrics compared to the losing site, oddly enough! Time on site, bounce rate for the losing site is much much better than that I see for the winning site. I checked many stats for this, not just GAnalytics, so I am very puzzled by this. # of thin pages are the same, some pages on the winning site are even thinner!

Some other info:
I've been building out my big site so I checked indexed pages lately and they were a mess. I had NOT NOINDEXed these pages under development, so they were included in the index. This could be a red flag. All in all, this update sure is confusing since I see many things wrong and many things right and am getting a lot of mixed signals.

I have a question though -- I wonder how much template navigation can play into this. One difference I see is that I had many more links on one site than the other. The big site is weighted down by more links to my product pages. The other site hardly has any.

[Content_ed]

@tedster

You're probably right. I was just going by my feel after reading all of the posts in that WebMaster Tools thread, which include a couple hundred legit sites at this point. But I didn't make a spreadsheet, so I probably remembered what I wanted.

I have been looking way out of the SEO box since I've never lived in that box to start with:-) One more thing the two unhappy sites had in common was abandoned Blogger blogs, since Google turned off FTP for Blogger last year. So I deBloggerized one of them, turned it into simple HTML pages with the same names and navigation.

But when I say the three sites are otherwise pretty identical, I really mean it. Designed in the same software with the same layout, similar numbers of navigational links. Perhaps less mission creep since it's relatively new, but most of the legit sites I've read about getting killed were single subject as well.

[Content_ed]

@falsepositive

My site that benifitted a little from Panda was also the smallest of the three, only draws around 700 to 800 unique visitors a day. It has no ads, no affiliate links, no community, no monetization. I get a lot of "sell me a link" requests for it.

But the next bigger site, a little over 2,000 uniques a day now but used to be 3,000, also has no ads, no community, maybe a tenth of the pages had a single Affiliate link for an online store that carrys a related product.

Both of the larger sites have many internal links to an order page, where the Affiliate link and some direct purchasing options are offered. The smallest site has none of that type of navigation as there's nothing for sale. But I'm doubtful that's an issue, since most of the sites I've seen that escaped Panda are covered with advertising, affiliate links, and internal order pages.

The smallest site had the fewest links on the home page since it has the least pages. Just for grins, I reduced the number of links on the homepage of the #2 site to be even less than the smallest site.

I do wonder if Panda has put a stress on the home page, something I was always indifferent to since I don't expect visitors to ever see our home pages. The small site has the highest text to links ratio of the three, but it's not 10:1 or anything like that.

[kd454]

I have 2 sites that were effected.

2 of them are almost identical in every aspect. 1 of the 2 got a penalty the other got about 30% increase in traffic.

Here are the differences between the two.

- They are in different niches

- The menu on the site that got an increase is tighter and less categories on the homepage menu. The menu on the site that got hit is very loose.

- The age of the site that got penalized is about 2 years, site that got the increase is 3 years.

- Much more incoming, recent links to the site that got penalized.

- The site that got penalized has a total of 80 pages, I went through the entire site it had 18 "thin" pages, content pages with 150-250 words, these pages were directly linked to from the homepage menu.

The site that got an increase has 130 pages, none are less than 400 words. And the pages are grouped into category's from the homepage menu unlike the site that took the hit.

- The site that got an increase actually has 3 ads units per page, the site that got hit only has 2.

I am thinking the reason for the site that got hit was, the 18 thin pages and the loose menu issue.

I removed/deleted the 18 pages on the site that got hit, replaced with 500+ pages original content. I will give this a month and if no change will tighten up the homepage menu and make it similar to the site that got an increase. Other than that there is nothing else I can do.

Both sites already have low bounce rate and 4 pages per visit avg.

What has happened since I removed the 18 pages from the site that took the hit is my CTR has tanked and income was cut in less than half from post Panda. Not sure why this happened but hope it recovers.

[tedster]

CTR has tanked and income was cut in less than half from post Panda

Adsense CTR, or CTR from organic rankings?

[Content_ed]

Did you mean 500+ words or 500+ pages? Hard to see where you could find 500+ pages of original content unless you've been writing for years and just never bothered uploading:-)

I have my doubts about 18 pages of thin content hurting a site. Many sites with nothing but thin content are doing fine or seeing gains.

[kd454]

@tedster

I was referring to Adsense CTR

@Content_ed
I deleted and replaced the 18 articles I removed with 18 500 word+ articles not 500 articles :)

The only reason I would think the 18 "thin" pages would be part of the issue is these all could have put into 1 or 2 articles. I have been meaning to either remove or group together for a while just never got around to it.

The fact that they were linked to directly from the main menu and really provided little for the user (poor user experience) could have pulled the trigger on the site.

The only other thing that is different from the site that got a nice increase is the format of the menu. The menu on the site that got an increase is set in categories, you click on the link and it take you to a page that has content and more links to other articles.

The site that took the hit has about 75% more menu links and each link only takes you to 1 article. Maybe they see the whole site as being "loose" since it is not grouped together in any specific order?

[Content_ed]

My baby bear site that took no hit only has around two dozen pages, so they are all linked from the home page.

The momma bear site that took the hit has well under a hundred pages (excluding a couple dozen blog posts), and most of them had been linked from the home page. After Panda, I changed that to just a half dozen links to category pages from the home page.

The poppa bear site has a mix of links on the home page, probably sixy or seventy all told, mainly to important pages or recent additions I wanted to give some juice.

[Content_ed]

OK, here's one more thing I hadn't thought of. My newest site is the only one that has Analytics on all the pages. The older sites don't because I never bothered. Wouldn't it be like Google to declare the presence of Analytics on every page an important sign of quality.

[TheMadScientist]

Wouldn't it be like Google to declare the presence of Analytics on every page an important sign of quality.

LOL, but No ... There are plenty of sites that rank fine without GA installed and it would be more like them to ignore the type of stat keeping you use.

[Content_ed]

@Mad,

I'm aware many big sites don't use Analytics. I would be surprised if they cared whether a site uses Analytics or not, it's using Analytics on just a fraction of pages that could be a crazy red flag. Google may have concluded that "quality" site use a uniform template, something like that.

[TheMadScientist]

That's interesting ... I don't use GA on my new ones either, which is why I think it doesn't really matter (and it would be pretty silly of them to count imo) ... It took a while, but I wrote my own stat scripts because I like to be able to organize my stuff better across multiple domains and do some things different than they do ... The 'consistent template' thought is pretty interesting though.