• http://twitter.com/brad_dalton Brad Dalton

    Search Engine Land remove all negative comments

  • http://twitter.com/jennyhalasz Jenny Halasz

    Hey Brad, what do you mean? I’d like to hear what you have to say. If you check my other articles you’ll see there’s often hefty debate. ;-)

  • http://wordswordsseowords.com/ Christopher Skyi

    “Duplication of a few pages, or a small percentage of poorly written content is unlikely to trigger Panda. The ratio usually needs to be pretty high.”

    Do you have any documentation to support that?

    My understanding is that “low-quality content on some parts of a website can impact the whole site’s rankings” (from http://googlewebmastercentral.blogspot.com/2011/05/more-guidance-on-building-high-quality.html).

    Google doesn’t say how much “some” is exactly. I would like if they were main pages for the site that had your main keywords, core pages if you will, then I would think those could trigger Panda, I doubt Google will actually give a ratio because then people will try to shoot for that ratio.

  • Deboti Chowdhury

    Thanks for the step by step tips…

  • http://twitter.com/marcusbowlerhat Marcus Miller

    There are some fairly easy ways to confirm or deny a panda problem and diving into your analytics and checking your landing pages is the first port of call. Look at the two weeks post a Panda update and compare to the two weeks prior and you will see the pages that have the most difference.

    Granted, some of these will be change naturally but if you see a clear pattern it can give you the first stage for some further research.

  • http://twitter.com/jennyhalasz Jenny Halasz

    Hey Christopher, that’s a great point and I should be clear that its purely based on my own experience. But after working with dozens of sites, that’s what our experience shows. You’re right that Google would never give a ratio. In my experience, the “tolerance” level varies by industry too. What my article is mostly trying to show is how to spot the warning signs before you’re affected by Panda, and how to know when and when not to worry.

  • http://twitter.com/jennyhalasz Jenny Halasz

    I couldn’t agree more! Well put. :)

  • http://www.brickmarketing.com/ Nick Stamoulis

    I worked with several clients who panicked when their traffic dropped (like we are all apt to do) but just a quick look under the hood made it fairly obvious that it was branded keywords that had actually taken the most substantial hit. You can’t make people search for your brand no matter how hard you try. In one client’s case their non-branded traffic had actually gone up a little bit but because so much of their traffic was branded it looked a lot worse than it really was.

  • Adam Berry

    Great post for my clients. It gives a very clear understanding of what Google does. Many thanks!!

  • k44g

    Can you comment on the possibility that syndicated RSS feeds might give rise to Panda “seeing” duplicate content, given the very idea of RSS syndication is to get the feed “out there” on other sites.

  • http://wordswordsseowords.com/ Christopher Skyi

    That would be more of a problem for the sites accepting the feed, if I understand your question. As long as the original pages on your site are using the rel canonical tag, you don’t have to worry. Also, for sites accepting feeds, most feeds just include the title and maybe meta description, not the whole article. As long as the whole site is not just a receptacle for rss feeds, snippets or entire articles, and is adding real value, I don’t think those sites have to worry about Panda.

  • http://wordswordsseowords.com/ Christopher Skyi

    It’s interesting when you say “tolerance” level varies by industry. I have one client who was hit by panda. I identified all the competitor sites and then used searchmetrics and semrush to see which had also been hit by Panda and which had not (based on the official algorithm update dates). I found most competitors where not hit by Panda and, not surprising, they had very strong sites. When I compared by clients main pages the these sites, her pages were clearly weaker, and there were only a handful of them. This makes sense. Panda rank orders a finite set of sites as a function of a query, e.g., plumbers NYC. So the relative comparison is between those those sites.

    So what you could be seeing a a set of sites that only start to significantly differentiate when you really step back and look at a lot of pages. In other cases, there can be a clear sharp difference between a hand full of pages. Looking a competitors who have NOT be hit by panda is very useful for understanding why a particular site got been hit.

  • k44g

    Your comment doesn’t bode well for aggregators — basically just a collection of feeds including some or all of the description element in addition to title — and which don’t add other content, quality or otherwise. But I think it logical to say that a webpage that (1) has quality content and (2) has a syndicated feed — say a half dozen of the most recent postings with partial descriptions (you know, with the “read more” link) — will not have trouble with Panda, on two counts: the quality and freshness of the content (assuming the new items are posted on the feed ‘regularly’). Oh yea, should mention that it’s important that the feed content be relevant to the theme of the webpage. Reasonable thinking?

  • Neil Forrest

    Its Pandamonium… or Pandageddon if you prefer.

  • http://twitter.com/jennyhalasz Jenny Halasz

    I completely agree with your observations; we have seen similar patterns.

  • http://twitter.com/jennyhalasz Jenny Halasz

    If you look at it from Google’s perspective, RSS feed based sites or aggregators don’t add any value to the internet beyond what other sites offer. You really have to have something unique to give Google a reason to list your site. I mean really it’s like that old adage concerned mothers give to their daughters… “why buy the cow, if he can get the milk for free”? Why would Google list the aggregator, if they can list the original source? At that point, I really do think it’s a percentage/ratio question. If 90% of your content is unoriginal, and 10% is original, you’ve probably got a problem. If it’s 50/50, maybe not so much. But again, I think it varies by industry, and I think something like a shopping site or a news site has a higher tolerance level than something like a biotech site. Again, this is all my opinion.

  • http://twitter.com/jennyhalasz Jenny Halasz

    Just want to clarify: As long as the original pages on your site are using the rel canonical tag **correctly**! You wouldn’t believe how many examples I’ve seen of it being implemented incorrectly.

  • k44g

    Whereas aggregators are touted as a way to get all your news in one place, rather than trying to track a large set of websites to tease out what’s new, I prefer to register for email notifications of new items on RSS feeds and/or getting those feeds in my Apple Mail account. Then my mail app is my aggregator in effect.

  • http://twitter.com/jennyhalasz Jenny Halasz

    Exactly. A straight aggregator with no additional content is just not valuable. And it’s not valuable to Google either.