homepage Welcome to WebmasterWorld Guest from 174.129.103.100
register, free tools, login, search, pro membership, help, library, announcements, recent posts, open posts,
Become a Pro Member

Home / Forums Index / Google / Google SEO News and Discussion
Forum Library, Charter, Moderators: Robert Charlton & aakk9999 & brotherhood of lan & goodroi

Google SEO News and Discussion Forum

    
E-commerce site - Panda recovery and lessons learned
arikgub




msg:4429512
 2:23 pm on Mar 15, 2012 (gmt 0)

One of my niche e-commerce sites got hit by Panda on Oct 14 losing over 75% of its traffic. Before that happened, the site basically controlled a small niche, holding #1 spot for all important keywords in the niche. After the Panda hit, all major keywords went to page 2.

In the past few days, my rankings are back to #1 spot, and it is a good time to sum up the recovery steps taken. Luckily, those were very few, so the situation is much easier to analyze and draw conclusions.

1. Root cause identified

Right from the start I suspected that the problem is with my product pages because of what Google may had perceived as a no-content template generated pages. I had over 5,000 product pages without "description". Product titles and images were the only thing that differed between the pages.

You might think that I was rightfully penalized. But before you judge my page the way Google did, consider this: I am selling a custom manufactured-by-request product, that is specified/described in great detail on autocad performed drawings. These descriptions appear on my product pages in a detailed high-resolution product image. It would be totally insane and comical to try to render these descriptions in text, the way Google likes it. In my view, my pages are rich in content and highly descriptive - it is just that Google is not smart enough to understand it. And that's what I was penalized for.

2. Actions taken

Only two changes (but substantial)

a. Applying the MC's advice to hubpages, I moved all my product pages to a sub-domain. The bots were banned from the sub-domain in robots.txt

b. The original taxonomy architecture was that the products were classified by a two-level categorization system. Every product belonged to some Category A => Sub category A.x. An additional change I introduced was to insert a canonical tag on all sub-categories specifying the parent category page as a canonical URL fro the sub-category page.

3. Result: changes in the index

All product pages and sub-category pages removed from the google index.

4. Result: changes in user experience

Nothing changed. Nothing. I haven't removed a single page on the site. I haven't even modified a single page on the site. The product pages are on the sub-domain now, and the sub-category pages have a canonical tag pointing to the parent category. But from the user's perspective - nothing changed.

Also, in that period I didn't add or delete anything from the site. Not a single page.

5. Result: rankings and traffic

Slowly climbed up the serps after the change. Got back on page 1 about a month after the changes were applied. From there continued gradual climb to the top of the page. In the past few days (5 months after the hit) got back #1 rankings for the most important keywords.

6. Summary

Assuming that the rankings were restored as a result of the applied changes, and not some other mysterious reason, the result is a good lesson for everyone who thinks that Panda is only about content quality or user experience. Sometimes it is just about Google being incompetent.

Remember, nothing changed from the user's perspective here.
Users haven't even noticed the change. And yet, it made google to go to such extremes as take the site down and back up to the first spot.

 

tedster




msg:4429536
 2:52 pm on Mar 15, 2012 (gmt 0)

Thanks very much for this report. It sounds like Panda really misfired in your case. I'm glad you were able to repair the damage.

Marketing Guy




msg:4429545
 2:56 pm on Mar 15, 2012 (gmt 0)

Very interesting, thanks for sharing.

Were any of the product pages bringing in referrals themselves prior to Panda? Have you seen any impact on sales, etc since recovery or was it all coming from the main keywords?

jqwan




msg:4429608
 4:26 pm on Mar 15, 2012 (gmt 0)

Interesting story but would be hard for most reg ecommerce sites to do as you would lose all value for all of your pages other then the home page and some cat pages since they are being no indexed and canonical. It would also stop them from any potential long tail searches most likely.

tedster




msg:4429618
 4:41 pm on Mar 15, 2012 (gmt 0)

One thing this report shows is that Panda is complex, and whether its demotion of a site is "deserved" or a false positive, getting to the root of it is far from one size fits all!

In this case, it underscores the limitations of web search today. We've known for a long time that information only present in images, or Flash movies, or even some Ajax pages, is problematic for all search engines. Google may have made some technical advances in these areas, but the publicity has been a lot stronger than the reality.

In s case like this present one, some alt attributes might have helped earlier on - as well as some on page text. Ideally a longdesc "should" be a solution too, but it's so underutilized right now on the web that I doubt it would have any ranking impact.

What I'm wondering about is why these pages USED TO rank - and what about Panda then tagged them as unworthy.

arikgub




msg:4429638
 5:16 pm on Mar 15, 2012 (gmt 0)

Were any of the product pages bringing in referrals themselves prior to Panda? Have you seen any impact on sales, etc since recovery or was it all coming from the main keywords?


Those product pages and sub-category pages were bringing some referrals. Not much, cause there was no much content to target the long tail anyway (as I mentioned, no "description" on the page). But yes, there was some price to pay by getting rid of those pages. That's why I am still a bit down in traffic and sales comparing to pre-Panda, even though the rankings for the main keywords returned.

But to be realistic, what I have now is probably the best I could do, I can't see how I can re-gain the long tail searches with the kind of product pages that I have. Google just can not handle them.

What I'm wondering about is why these pages USED TO rank

tedster,

As I said in my reply to Marketing Guy the product pages didn't rank particularly well even before Panda, but at least that didn't hurt the rankings of other pages. With the advent of Panda, these "low quality" pages dragged the entire site down, including the home page which used to rank #1 for many relevant keywords.

Marketing Guy




msg:4429641
 5:23 pm on Mar 15, 2012 (gmt 0)

It is worrying for the industry that this kind of thing is becoming such a problem. Back in the day there might just be some vague concerns about indexing quotas or PR distribution if you have a bunch of non-value pages indexed (forum profiles, blog categories & tags, ecom site product variations and so on).

A shift to the point where these (actually quite legitimate and common CMS setups) can kill off your rankings entirely is actually quite an aggressive move by Google. The question whether or not it's intentional (clean up your low content pages or get out), or not (oops we thought you were a content farm)?

I can see arguements either way (it'd put a stop to forum profile spamming as a link building technique for example).

Global Options:
 top home search open messages active posts  
 

Home / Forums Index / Google / Google SEO News and Discussion
rss feed

All trademarks and copyrights held by respective owners. Member comments are owned by the poster.
Home ¦ Free Tools ¦ Terms of Service ¦ Privacy Policy ¦ Report Problem ¦ About ¦ Library ¦ Newsletter
WebmasterWorld is a Developer Shed Community owned by Jim Boykin.
© Webmaster World 1996-2014 all rights reserved