Just got this email from Kenneth from Ezarticlelink. I used to use his service but I think it's pretty much past it now so I don't recommend it in anyway. He's got some interesting thoughts about Penguin and it's well worth reading. Just bear in mind that he's basically saying that it's still OK to use Ezarticlelink and that his evidence and point of view about Penguin may be a little biased! It's an interesting read anyway: Google Penguin Update - How Now? 1) What is Google Penguin Update? On 24/4/2012, Google launched a major algorithm update called webspam update, later named as Penguin update. The objective of the update is twofold. i) To weed out spam contents such as keyword stuffing and poorly written articles such as those unreadable spun articles (I thought they have done it in Panda update?). ii) To weed out sites that build tons of spam links to boost their search engine ranking. Do not confuse Penguin update with Panda update, which is another update done last year, mainly targeting at website quality. We were unscathed by Panda update, but Penguin update has caused many of our members lose their rankings. But is the blame on us, or they are just victims of the disaster that has affected both black-hat and white-hat SEOers? We'll see. 2) Can Penguin really detect link building? Something that smells fishy to me is, if Google were to launch a webspam update that can flag sites that build links, why does it take another effort to kill blog networks like BMR in March? For those who don't know what I'm talking about, BMR is a very popular private blog network that allows its members to post articles and build backlinks. The network was de-indexed on 22 March, one month before the Penguin update. I may be wrong but here is my guess. Penguin can't detect link building activities. What it can detect is links from the same or similar articles, using Google's LSI technology (it doesn't have too many new technologies, does it?). So when Penguin sees too many backlinks from essentially the same or similar content, those links got flagged. If so, the new update will not have any impact on networks like BMR, because BMR syndicates ONLY unique article, and every article is only published ONCE. If Google were to launch Penguin without killing BMR, BMR will be happily singing to the whole world 'you cannot catch me, you cannot catch me!'. This will put Google in a bad light. To avoid that, Google needs to kill BMR before it launched Penguin. 3) Can Penguin really detect links from poor content or irrelevant content? I'll show you proof that it can't. But that doesn't mean you can start submitting junk articles to our network because our Human Reviewer will not allow this to happen! Case Study - Roofing Toronto Do a search for roofing toronto and you will see the top result is from a website called Professional Roofers. A search in href.com reveals that it only has 65 backlinks. Are these high quality and relevant backlinks? We'll see. Below is a "typical" backlink it has. It is a link from an article called The Delicious Kiss. Huh, what's that? Another link from a completely irrelevant article. Here is the ultimate. Read the content. It will make you laugh. You can do your own research. Most of its backlinks are junk links like these. But there is one thing that makes all the difference. This site has a link from business.com directory, which has a PageRank of 3 (and business.com itself has a PageRank of 6). Notice that the anchor text listed in business.com directory is Professional Roofers, not Roofers Toronto. The reason why this website can rank high in Roofers Toronto is because of the spam links it has built because the anchor text for those spam links are either Roofers Toronto or Roofers In Toronto. So here's what's happening. Business.com boosts the authority of the site. With that authority, this site can now rank high in the keyword it wants with a slight boost in backlinks, eventhough those backlinks are pure junks. And for your info, these junk articles all have 0 or unranked PageRank. As we can see, you don't need a lot of backlinks to rank well now. What you need is authority, and some backlinks which you have control over the anchor text. 4) Can over-posting of articles hurt my ranking? This is a question I've been asking myself after the penguin update and I thought it will be so, since Penguin is about killing webspam. That is why soon after the penguin update, I reduced our default posting limit to 50 and slow down the default posting speed to 0-5 articles a day. But am I right? Look at this screenshot of our keyword manager for a domain that is still doing well: Keywords from another domain that are still ranking #1 for all 3 keywords: Look at the second keyword in the screenshot below. It has 7 articles and 2691 posts. Still good. Screenshot from another domain showing the same. This keyword below has 529 posts from a single article. Still doing well. Finally the ultimate one, 708 posts from a single article. Rank #2! (But note that this is a Squidoo page. Perhaps it can take more stress or it really doesn't matter?) Note that I only use internal pages for investigation. Homepage will usually do better and it's not a good example. These internal pages only use our network to build backlinks, so they are a better gauge. 5) If nothing is wrong, why do I lose all my rankings? I didn't say nothing is wrong. The screenshots above are meant to show you over-posting will not hurt you. We do not know what is Penguin's threshold. Perhaps it is a combination of different things that cause links to be flagged. In fact, your rank drop may have nothing to do with spam links. With the new Penguin algorithm, it is obvious that Google now put more weight on site authority (i.e. site PageRank). Internal pages of authority sites are now dominating Google's search results, eventhough they may be less relevant, with little or no backlinks! 6) A special case where EzArticleLink can REALLY hurt you! This happens to a few members and I believe this is a confirmed case. Here is what happen: Say we want to build links to url A. What some members did was they take the article in url A, spin it and submit to our network. (I always advise members not to do it. Articles that you want to rank should always be 100% unique). In the past, they manage to get url A to rank high. But after Penguin update, their ranks drop flat. Remember I said that Penguin's objective is twofold? Besides devaluing spam links, it also devalues spam contents. When you publish spun versions of the original articles on your website, Google may mistaken your original article as spam content and hence it devalues it. During my investigation, I came across an interesting case. I found 2 members who build links to their Ezinearticles.com articles using ONLY ezArticleLink. One member has lots of his Ezinearticles.com articles rank high in Google. Another member has lots of his Ezinearticles.com articles disappearing from the top 100. Even if some articles manage to squeeze into the top 100, the rankings are lower than other Ezinearticles.com articles with fewer backlinks and even lower than the articles he has posted in our network! Here is a typical example I saw: For a golf related keyword, Google position #13, another article from Ezinearticles.com. Total backlinks only 4, from 2 IPs Google position #16, #19, #21, #22, spun article that THIS MEMBER posted in our network! Google position #23, this member's article in Ezinearticles.com After considering all possibilities, I finally realize that this member submitted spun version of his Ezinearticles.com articles to our network whereas the member who did well used a different article. Another member who practiced the same strategy on his own website also experience the same thing. So please, make sure that the article you want to rank in Google is 100% unique. Do not publish that article elsewhere, even if it is spun. 7) What are the changes we've made in response to Penguin Update? We made these changes soon after Penguin update. If you've been submitting articles recently, you should have noticed the change. - We now allow members to decide their initial posting limit. The default initial limit is 50, with posting rate of 0-5 articles per day. - Beyond the initial limit, posting rate will be reduced to 0-2 per day (Just to maintain links that are lost) - Beyond 300 posts, posting stops. (This to prevent over-distribution, though we are not sure whether there's really an issue of over-distribution.) The changes apply to both new and old articles. Please note that the new limit is adjustable. If you prefer to post an article 300 times like the old days, you can still do so. Just change the initial posting limit to 300 (or 150 for silver members). We will revise these measures as we conduct more testing. But for now, I think it's better to be conservative. 8) What are our future plans? Moving forward, our plan is to make your articles more unique (but without you spinning it more!). This plan will take some time to implement because a lot of programming and testing will be needed. We also plan to increase the keyword tags in your article (i.e. instead of 3 keywords, you may have 6), so that you can have more variety in your anchor texts. Note that these are plans, not promise. If something is good for our members, we will do it. But halfway through the implementation, if we feel that something is not right, we will pull the brake. 9) What should I do now? Penguin is a change in algorithm, not a penalty. If you suffer a rank drop on 24/4/2012, then it is due to Penguin update. If you suffer a rank drop before 24/4/2012, then it has nothing to do with Penguin. If you are affected by Penguin update, there is no need to delete the articles you've posted in our network (unless those articles are spun versions of the original article on your site, as explained in #6). The worse case is those links are not counted. As what I've shown you in the screenshots above in #4, even a single article with 700 posts can make a url rank high in Google, it gets to show that either those links are still counted, or they are merely discounted. They definitely won't hinder a url from ranking well for a particular keyword. What you need is more backlinks from unique contents, with varying anchor texts and from varying sources. You also need backlinks from high PageRank webpage (not website) so as to boost your authority level. Those backlinks can be through any anchor text. It doesn't matter. What you want is the PageRank juice (See #3 for case study). In fact, the reason why your rank drop could be simply because you have a lower site authority than your competitors. Trust me. Google Penguin still can't differentiate between built backlinks and organic backlinks. What it can detect, at most, are links from similar content. So make sure that your articles are well spun, especially the title and the sentences around the keyword tags. Also, don't just build links from our network. You need a mixture of backlinks, from social bookmarks, Blogspot, Squidoo, Ezinarticles, Myspace, Facebook, Tumblr, blog comments, forum links, a mix of nofollow and ******** links, links from authority web directories etc. I know it's tedious but you need to do it at least for your homepage in order to increase its authority level. Lastly, if you have Bonuslink credits in your account, use them! The good thing about Bonuslink is you can insert as many anchor texts as you like and the links are posted on different articles in the same niche. So naturally, you are getting backlinks from unique, relevant content with varying anchor texts. Voila! To prove its effectiveness, here are some screenshots I took at the time of writing (9 May 2012). For the privacy of this member, I need to hide some of the keywords. But you can briefly see that they are very different keywords. Can this member rank his single url for these varying keywords? Let's see. For the first anchor text (*** supplies), The url ranks on page 2. Note that this anchor text is found in its title tag. So there is an onpage advantage. Let's look at another anchor text (tool ***). The same url ranks #2 even though there is no exact match in the title or within the content. Same goes for another anchor text (*** workwear). The url ranks on page 2 even though there is no exact match in the title or within the content. That's the power of our Bonuslink, so use it! Before you get overly excited over Bonuslink, please understand that this site has a PageRank of 3. So it already has some form of authority. As what we've seen in the Roofers Toronto case study, with authority, all you need is some anchor-text-focused backlinks and you can rank for the keywords you want. Last but not least, let me end this discussion with the screenshot below: This is the same search term that we used when we rewrote our homepage after Panda update last year. You can go to our homepage to see, or you can see from the screenshot below. We used to dominate 10 out of 10 listings for this search term. Now we still have 6 out of 10 listings. Based on my latest statistic, our entire article network still receives slightly more than 2 million traffic a day! Forget about link building. Focus on submitting articles that benefit people. Some of the traffic to your article may eventually go to your linking site. Hey, isn't that what Google suggests? Promote your website to get traffic, not to build links for SEO.