Jonathan Leger – SEO And Internet Marketing Blog Internet Marketing Blog

24May/12Off

Anchor Text Optimization – How Much Is Too Much?

In my last blog post I talked about how to rank in Google after their Penguin update. One of the things I pointed out is that Google is penalizing sites for "anchor text over-optimization." That is, if you get too many links with the exact keywords you want to rank for in the anchor text, you might get penalized.

Notice that I said you might get penalized. That's because there are some circumstances in which site's aren't being penalized. More on that in a bit.

After reading about Google going after over-optimized sites, I decided to do some data mining and see if I could figure out exactly how much exact-match anchor text is too much. Here's what I did:

1. I gathered 1,500 keywords from 30 very diverse markets. Everything from business to technology to transportation to food and beverage to chemicals. A wide range of markets.

2. I ran those keywords through Google to get all of the top-level domains ranking on the first page for the keywords. I removed all ranking inner page urls from the data set.

3. I checked the link profiles of all of those domains to see what percentage of their links had anchor text matching the keywords they were ranking for.

Before we continue, let me explain why I removed the inner page urls from the data set. As I also pointed out in my previous blog post about Penguin, Google is favoring authority ("big box") sites in their search results more highly than ever. That means that an authority site can get an inner page ranked with very little or no backlinks at all. Wikipedia is a big example, as is Amazon.com. They're all over the search results after Penguin, as are other authority sites.

Because those authority sites would skew the results of my data mining (since they would typically have tiny exact-match anchor text percentages), and because I'm pretty sure most of my blog readers don't run huge authority sites, I only logged the data for the top-level domains whose home pages were ranking for the keywords. That is, if mydomain.com was on page one, I kept it, but if mydomain.com/innerpage.html was ranking, I dropped it. So the results you're seeing reflect sites that are more typical of what the average webmaster would be able to achieve.

Okay, onto the numbers. Some of what I learned was quite enlightening.

(For the purposes of this blog post, I'll refer to the percentage of links whose anchor text exactly match the keywords the site is ranking for as its EMA. Also, all of the queries were done via Google.com, so the ranking sites generally favor the USA. Lastly, all of the linking data was gathered using KeywordCanine.com.)

 
1. The Average EMA Is Pretty Low

It probably won't come as much of a surprise for me to tell you that the average EMA for a site is pretty low -- just 10% across all of the markets. So, on average, only 10% of the links to a ranking site contain the exact keywords the site is ranking for. But don't take that as the standard to aim for, because it varies a lot across markets.

Here's the full list of the markets, their average EMA, and the Maximum EMA found for any site that ranks for its keywords in that market. For the Maximum EMA, only sites with at least 50 unique domains linking to it were considered.

Average anchor text diversity across topics: 10%

Market Avg EMA Max EMA
business 9% 97%
home based business 17% 57%
work from home 14% 53%
b2b 4% 22%
business management 5% 25%
b2b ecommerce 18% 18%
management 9% 49%
office management 4% 8%
online stores 36% 36%
semiconductors 4% 15%
software 5% 60%
startup 9% 33%
technology 6% 80%
transportation 3% 26%
virtual server hosting 8% 11%
web design 18% 98%
web hosting 8% 62%
food and beverage 8% 8%
office supplies 2% 36%
biotechnology 19% 42%
email hosting 15% 26%
office backup 6% 10%
plumbing 29% 92%
agriculture 4% 44%
pharmaceutical 9% 60%
human resources 12% 40%
internet security 1% 16%
chemicals 6% 93%
ecommerce 4% 41%
aerospace 6% 54%

From these figures you can see that the average EMAs are really low, but that there are usually sites with a much higher EMA than the average ranking on page one for their keywords. So who's getting away with having a higher EMA, and how are they doing it? Read on.

 
2. The Get Out Of Jail Free Card - Exact Match Domain Names

In pretty much every market I tested, Google is ranking one or more exact-match domain (EMD). What I mean by "exact match domain" is a domain whose name is made up of all of the terms in the keywords it's ranking for. That is, if workfromhomeonline.us is ranking for the keywords "work from home online", that's an exact match domain for the keywords. That site is ranking for those keywords, by the way, despite having a 53% EMA.

However (and this discovery goes against much of what's said about exact match domains), domain names that have dashes in between the terms also appear to be benefiting from this exception to the over-optimization penalty. That is, work-from-home.biz is ranking on page one for "work from home" despite having a 70% EMA.

Those two examples also bring to light something else that goes against common SEO knowledge -- Google is not penalizing EMDs even if they aren't one of the "big four": .com, .net, .org and .edu. The "lesser" domain name extensions are also being exempted: .info, .biz, .us, .ie, .ro, etc. They all appear to be averting punishment despite having very high EMAs.

Here's a list of some of the exact-match domains ranking in Google in the markets I tested. Each of the ones in the list have a 40% EMA or higher:

Keywords Rank* URL EMA
internet business expert 6 http://internetbusinessexpert.co/ 97%
work from home 4 http://www.work-from-home.biz/ 70%
work from home online 2 http://workfromhomeonline.us/ 53%
work from home india 6 http://www.workfromhomeindia.biz/ 50%
work from home ideas 6 http://www.workfromhomeideas.us/ 67%
b2b strategy 5 http://b2b-strategy.ro/ 42%
business management solutions 6 http://www.businessmanagementsolutions.ie/ 100%
pain management 10 http://painmanagement.com/ 72%
music management 7 http://www.musicmanagement.com/ 49%
music management software 8 http://www.music-management-software.com/ 100%
network management software 3 http://www.networkmanagementsoftware.com/ 40%
obsolete semiconductors 2 http://www.obsoletesemiconductors.net/ 100%
technology new 8 http://www.technologynew.org/ 86%
computer technology 10 http://www.computertechnology.com/ 67%
la crosse technology 1 http://www.lacrossetechnology.com/ 80%
technology books 4 http://www.technologybooks.com/ 55%
creative technology 3 http://www.creativetechnology.com/ 58%
web design company 1 http://www.webdesigncompany.net/ 88%
affordable web design 5 http://www.affordablewebdesign.net/ 53%
new web design 1 http://newwebdesign.com/ 57%
top web hosting 8 http://topwebhosting.com/ 50%
web hosting review 6 http://www.web-hosting-review.net/ 41%
web hosting canada 2 http://webhostingcanada.org/ 51%
food beverage canada 2 http://www.foodbeveragecanada.com/ 48%
biotechnology companies 4 http://www.biotechnologycompanies.net/ 50%
exchange email hosting 6 http://www.exchangeemailhosting.com/ 100%
plumbing contractors 4 http://www.plumbingcontractors.com/ 56%
plumbing how to 6 http://www.plumbinghowto.biz/ 100%
home improvements maryland 10 http://homeimprovementsmaryland.net/ 100%
pharmaceutical jobs 6 http://www.pharmaceutical-jobs.com/ 100%
pharmaceutical jobs 8 http://pharmaceutical-jobs.org/ 60%
pharmaceutical technology 3 http://www.pharmaceutical-technology.com/ 48%
computer internet security 8 http://computerinternetsecurity.org/ 92%
pharmaceutical chemicals 3 http://www.pharmaceuticalchemicals.com/ 100%
ecommerce marketing 6 http://www.ecommerce-marketing.com/ 67%
bigelow aerospace 1 http://bigelowaerospace.com/ 54%
plumbing supplies 9 http://www.plumbingsupply.com/ 92%

* Due to Google's localization, the rank you see the site at may be different than what's shown in the table.

Just look at those EMAs! Google is clearly letting those domains get past any anchor text over-optimization penalty. Also notice the last entry ("plumbing supplies"). Google is giving a pass to plumbingsupply.com even though its exact match anchor text is for "plumbing supplies", not "plumbing supply." So it seems Google is also exempting exact match domains which it determines have a variation of the keywords in the domain (in this case "supply" instead of "supplies").

So why is Google letting these guys get by without a penalty? It makes sense, really.

If Google penalized sites with a high EMA even if their domain name was an exact match for the keywords, they would end up dropping all kinds of brand-name sites out of the rankings. Think about it: what anchor text do most people use when linking out to a brand-name site (like Ford or Adobe or Amazon, etc.)? Their name, of course! So Google has to give those sites a pass despite having a high EMA. It just makes sense that Google won't penalize you for links with anchor text matching the keywords in your domain name. That's often the site's brand, and it will naturally have a much higher EMA for those keywords.

Is Google giving a pass to all EMDs with very high EMAs? The data can't answer that question. But clearly they are giving a pass to a lot of them.

 
3. The Ranking Exact-Match Domains Have A Lot Less Links

Another important point about the ranking exact-match domains versus all of the other ranking top-level domains: they have a lot fewer links aimed at them. In fact, on average the EMDs only have about 15% as many links as the other ranking sites.

One stand-out example is pharmaceutical-jobs.com, which is ranking on page one for (of course) "pharmaceutical jobs." It only has about two dozen external domains linking to the entire site. The other top ranking results typically have many hundreds or thousands of domains linking to them. Clearly Google is highly favoring the EMD in this case.

Here's a breakdown of the number of links from unique external domains ranking the non-EMD sites versus the EMD sites:

Market Avg Links Avg EMD Links  
business 1,901 1,556 82%
home based business 156 225 69% more
work from home 204 85 42%
b2b 289 23 8%
business management 70 1 1%
management 350 79 22%
office management 138 6 4%
semiconductors 644 2 0%
software 1,438 417 29%
startup 135 64 47%
technology 414 115 28%
transportation 1,166 121 10%
web design 435 475 92% more
web hosting 2,949 159 5%
food and beverage 430 40 9%
office supplies 371 40 11%
biotechnology 72 2 3%
email hosting 143 278 51% more
office backup 70 70 100%
plumbing 502 5 1%
agriculture 1,232 595 48%
home improvements 970 1 0%
pharmaceutical 243 135 55%
human resources 260 68 26%
internet security 930 11 1%
chemicals 273 35 13%
ecommerce 2,835 745 26%
aerospace 316 137 43%

 
Some Non-EMD Sites Are Also Getting Away With It

The data also shows other sites with very high EMA values getting a pass from Google even if they don't have an exact-match domain name and aren't a brand. Why Google is giving those sites a pass isn't clear. For example, addonchat.com is ranking for "chat software" with a 60% EMA, and plimun.com is ranking for "web design" with a whopping 98% EMA. If I figure out why Google is letting these sites get away with that, I'll definitely be blogging about that, too.

 
The Take-Away

So what can you take away from all of this data and these numbers? In short, exact-match domain names are your friend! They can be ranked with a lot fewer links and apparently have a much better chance of not getting penalized for anchor-text over-optimization. This includes the "lesser" domains (.info, .biz, etc.), as well as domain names with the keywords separated by dashes (e.g. work-from-home.biz).

Also, if you're not using EMDs, it's important to diversify your anchor text a lot. How much is "a lot" really depends on your market. So do the research. Check out the link profiles of other ranking sites in your market and see what their anchor text looks like.

If you have any questions or comments, or would like to suggest other post-Penguin ranking factors for me to dig into in a blog post, please leave a comment below. Your feedback is always welcome!

Oh, and one last note: if you found this post beneficial, please share it using one of the buttons below:

Related Internet Marketing Q&A

  1. Is It Important To Vary The Anchor Text For Tier 2 And Tier 3 Backlinks?
    7 Answers Available - Search Engine Optimization - Asked 338 days ago

    What do you thinks? Is it going to affect the result? Anyone tested it?

  2. Do You Prefer Text Based Training or Video Based Training?
    35 Answers Available - Internet Marketing - Asked 357 days ago

    I am working on a new training course for online marketing and I want to know how you like it. For learning anything on the net how do you prefer your training and why? Assume all of the below are step-by-step.

  3. How can I make my backlinks "safe" in Google's eyes?
    10 Answers Available - Search Engine Optimization - Asked 384 days ago

    The Facts About Link Building We know that Google says that you should never buy links to help your ranking, or build links in any "unnatural" way. Everything should be "organic" and "natural", blah blah blah. But we also all know that very,...

  4. How to avoid on-page SEO over-optimization?
    3 Answers Available - Search Engine Optimization - Asked 379 days ago

    Some years ago, we needed implement ALL of SEO rules to make more money but, now, over-optimization + bad content can penalize your websites. So, how do you avoid over-optimize your on-page SEO? I only check if keyword is present on title and H2 tag...

  5. How can I put my advertisement on Internet Marketing Q&A?
    14 Answers Available - Miscellaneous - Asked 379 days ago

    Want To Put Your Ad In Front Of Thousands Of Visitors A Month For FREE? You can, and it's easy! Just register for an account here at Internet Marketing Q&A. Then be an active member by answering people's questions, asking your own,...

Comments (171) Trackbacks (0)
  1. Once again Jonathan,

    Insightful knowledge you’ve provided,

    Happy to be on your email list.

    ED

  2. Wow, great research, this really provides some perspective and know that I think about it, a lot of my exact match domains were less affected by the update.

    Cheers

    Phil

  3. Great article and thanks for sharing all the research you did. I’m sure you didn’t compile it all over morning coffee! = )

    I could be wrong but I do believe that Panda reduced the importance of EMD’s but after Penguin it seems they are back as my local market has seen a surge in EMD’s rising back to the top of hyper-local search results.

    With this being said, sub-domains are your friends guys! I’ve been using them almost exclusively for local clients for over two years now and zero sites were effected after Penguin. That’s not to say the sub-domain in and of itself held all the ranking power. We still need great content and good onsite optimization as well.

  4. Thanks for the blog post. I am trying to read as much as possible on the Penguin subject. All that you say has been taught for years. The domain name matching your key search term I took advantage of that term. {snip} has always been easy to be on page one for that term. Also the hyphens between promoted a long time too. That fellow in Canada (I forget who) with his pillars, thinks it is more important for his customers to have the dash marks than to miss out on the proper keyword.

    That is good advice to use these two methods. The problem is that I think people get embarrassed using either. I recently built 10 blogs to get more adsense profits using your article machine. I was going to try to use another persons help with some aspects that caved in. So I decided to sell the domains on ebay. The ones with hyphens, even though they were perfect key word matches sold for less than a dollar. The ones with no hyphens sold for a slight profit. THe ones that sell for most money are shorter, dont match anything. The shorter the better.

  5. Hi Jonathan,

    I manage several websites and am trying to learn how to get my YouTube videos on Google’s first page.

    I tried owning the actual domain and I found that Google did in fact give me first page ranking for a website.

    Now, I want to use one domain, such as my phone number domain: http://www.123-456-1234.com and, when I make a video, and upload it to YouTube, I will include in THE EXTENSION, my keywords, in the description of the video… such as, for the keyword “guaranteed free teeth whitening” …

    http://www.123-456-1234.com/guaranteed-free-teeth-whitening.html

    Followed by some description text…

    My question is: Do you have any data for the value within Google of having the keywords in the extension????

    It seems to me that it does help.

    Thanks,

    John Burch

  6. John, great post and analysis. Im still trying to do the same and get down to the nitty gritty of what Penguin has brought other than some poor shoddy results and a lot of termoil with the spam community.

    I agree with your data and it makes sense but I have a few examples with exact match domains with lots of anchor text for the keyword in the domain and they got heavily peanalized. This goes againts your data obviously. Some of them are affiliate sites though which maybe one reason.

    One thing I have found also here in the UK is that websites that got peanlized from penguin were also ones with a lot of links from .com sources. The ones with majority .co.uk links survived and excelled. Maybe thats because there .com links were spammy? Not sure. I still need to do some digging on the backlink profiles to be sure.

    Any how, good stuff dude. I am fan :-)

  7. Excellent article.

    Link building is going to be very different from now on. There have been signs that link variety was going to happen and now we have more evidence that it is actually here. I am just glad that I have 10 different links for each page that I create. I think that it is important to stay below 10% when building backlinks. Yes, a higher percentage may work for now, but what about in 6 months or even a year?

    Adam

  8. Thank you so much Jonathan!

    I have been following Google Penguin updates from all kinds of sources as I lost 90% of my traffic, in the disaster – just when my business was exploding! Funny :-) !

    The information and testing you provide is the best on the topic for me even though I am really on the subject!

    Thank You!

    P.S. Now I have a complete knowledge (including my own research) to go the battle again and fix my traffic probs!

  9. Show me any website that isn’t an authority site( or at least a high PR site with a lot of traffic) but has lots of links pointing to it and I’ll show you a very busy webmaster who is spending a lot of time “building” links to his site (or paying someone else to do the work).
    I know that, you know that–everyone knows that including the search engines. Yet links are still considered the main basis for ranking a site. Why? They have absolutely no correlation to the value of the site–they only show how hard someone has worked to insert links into websites all over the internet. I would include social media backlinks in this also. The vast majority of them are placed by website owners–not by third parties–again unless they are paid to do so. Tweets and likes–buy them on Fiverr. Links from article directories–write your own articles or pay someone to write, spin and submit articles to hundreds of directories. It is all quite meaningless as far as the real value of any website goes. The winner is simply he who buys or creates the most links back to his website. Companies hire people whose sole job is to promote their website by link building and related work.
    So whats all the fuss about diversity of anchor text, exact match domains, domain extensions etc. Does it really make any difference if the guy who is building all these links uses the same anchor text all the time or varies it widely. Does it make any difference at all to the value of the website if the domain name is an exact match to the main keyword. Yes I know “it looks more natural” if there is a diversity of anchor text. So lets go on pretending that all those links are “natural” and not built by the website owner or his paid minions.
    Every few months (or weeks) lets decide that some types of links count for less than they used to (or lets be really nasty and penalise websites with these links which used to be the cream of the crop). We all know that forum profile links have been used to the max in the past, and the search engines gave them lots of power. Now they are at best useless and at worst may incurr a penalty. Web 2.0 links still are still acceptable at least– but are they really any different? Of course not.
    Personally after having some high ranking websites drop right off the map because of changing ” algorithms” I have decided to drop right out of all internet efforts that rely on this ever changing whim of the day, and to derive my traffic from elsewhere. It is simply not possible to run a business which at one moment is ranked highly because it complies with the “algorithm” only to have that set of rules change overnight and be consigned to the netherworld never to be seen again. JM.

    • Well said, sir. I agree with you that links have long since ceased to be a real measure of the value of a page or a site. That said, Google is still the 800lb gorilla, and ranking in Google can deliver enormous amounts of traffic. As long as that is the case then analyzing how Google ranks pages will be an important part of many webmaster’s business model. It should -not- be the only thing you focus on, though, as I detailed in a previous post.

      In Google’s defense, however, we’re left with the question: if not links, then what? If all on-page factors are equal, how does a search engine determine which piece of content of equal metrical value should be displayed first? They are experimenting with social factors, “stickiness”, authorship, etc. Then there are the alternatives like Blekko that have the philosophy that metrics simply aren’t enough — you have to have people at least partially responsible for deciding what gets displayed. Is Blekko better at returning relevant results than Google? Your call.

      So what’s next? Which direction should the search engines go? Am I coming up with a good idea for my next blog post? ;)

      At any rate, I wish you all success in divorcing Google. If it’s causing you too much stress it’s best to just leave it be. There’s plenty of traffic to be had from other, more stable, sources.

  10. EMD seems to be an advantage, however, it’s not a guarantee that will protect your site from being devaluated – let’s not forget that the content on the site still is king. In some niches you can build a site with good content (and admittedly EMD) and almost without any backlinks and still get on GP one for niche-specific keywords.

  11. Excellent research! Thanks for digging into this. Great to hear the the staple of our web marketing strategy, EMD and keyword linking is still a viable way to achieve ranking – if done correctly. Well done!

  12. From a MUCH smaller sample size (I monitor about a dozen of our sites), this data mirrors what I’ve seen as well. Nice to get some real data to backup what I’m seeing. THANKS!

  13. This is a really great post Jon, amongst all the hyperbole with the Google updates it’s brilliant to see someone talking sense and showing how it has actually affected the rankings.

    We use a number of exact match domains and although we add a blog to each one for fresh content, there hasn’t, generally, been a heavy push for building links, and at the time of writing, not one has been affected by the updates.

    It’s good to know that it isn’t a fluke and that this strategy, for the time being, hasn’t been penalised.

    Have a good day!

    Cheers

    Leon

  14. Thanks for posting this. Can’t add much from my experience although I have a few experiments going at the moment so will be interesting to see if they match your results.

  15. Interestung research, but I’m not sure about the conclusion. I’ve had several EMDs blasted into oblivion by Penguin.

    I notice that the examples all have a low level of backlinks. The conclusion shouldn’t necessary be that this means their EMD carries a huge amont of weight. Can’t it just be that they’ve risen to the top as all the over-optimized sites have gone?

    Google have stated before that their strategy for getting the best sites to the top of the search results is to remove the spam. So, rather than giving them a push, I think they’ve likely just naturally risen up without a bump at all.

    Also, I believe Matt Cutts has acknowledged in the past that EMDs did get an unfair ranking boost and that they were going to stop that. Seems strange that they’d change their mind on this point, especially if EMDs have been used for manipulation and ranked the kind of sites that Google doesn’t like in the past.

    So, I see your point, but I’d bet that those EMDs will be easily beaten in the rankings by SEOs who adapt to the over optimization stuff, and start building authority in the ways that those big brands that you mentioned have.

    Having said all that, it’s all a bit crazy at the moment so who knows what’s possible!

    • Many of the examples have hundreds of backlinks — one even has 1,500+. And that’s how many unique domains are linking to them.

      The point is that these EMDs are doing what Google has claimed they were penalizing in Penguin — over-optimizing their anchor text — and yet they are ranking well. They may have just “risen to the top” when the others were removed, but the point is that based on what Penguin was supposed to do these sites should have been removed too. But they weren’t, and it seems that they weren’t because of the EMD / anchor-text correlation.

  16. Hi Jon
    I paid to have a website built, a couple of years ago, to apply to be an ebay seller. It didn’t work, and I haven’t touched it at all.

    It now ranks at number 3 on google USA, in a fitness keyword with 231 million results.

    I assume it is because it has an EMD (.net), and has no backlinks, therefore hasn’t broken any rules.

    Sites that previously ranked above it have probably fallen away because of penguin or panda.

    If I knew what I was doing, I’d monetise it.

    Great post. Thank you.

  17. Hey Jon. Excellent post and… wow, your research, as always, is impeccable.

    This kinda fits in with my own theories about where Google is going with all this. They’ve actually been paying attention – apparently close attention at that.

    For years ALL of the experts have been hyping EMDs with a .com, .org, or .net extension. It has been at the core for almost every single Adsense or Amazon course out there. And, you know what. Those EMDs for the big 3 were gone a LONG time ago – as well as almost every reasonable variation. So, a lot of folks were “stuck” with looking elsewhere – .biz, .info, .cc, and so on.

    Google, in their infinite wisdom (sic) probably figured those people, by default, are not as “SEO savvy” as the gurus they were learning from. So, they decided it was an easy way to sift the novices from the experts.

    It’s a theory, anyway.

    I dunno where this is going. But it is interesting.

  18. It’s all about Good Old White Hat SEO. Create good content like this blog.

  19. This article starts with the premise that EMA has effected the results but shows no data about the rankings of sites before the update so I am going to say that the premise of the whole article is not backed up by anything. There is also no evidence of any penalty for high ema since we dont know where these sites ranked before and we dont know what sort of links they have or what sort of links the ones that dropped had or anything at all really. If sites did drop assuming that some did we don’t know from this data that they were penalised. Probably their links were devalued but that is not a penalty it just means other sites have better links. Same thing happened when reciprocal links were discounted and that wasnt a penalty either.

  20. how does your 3 way linking system hold up to all the changes at google.
    have the sites maintained there rankings?

    • Still working fine for my sites. Just make sure you either use the spinning feature of the anchor texts or you make sure you have a diverse link profile already to prevent 3WL from raising your EMA too high.

  21. Thanks for the handy research Jon.

    I tend to choose EMD names as much as possible, but it is sometimes hard to find what I want – obviously a couple of other people have the same idea.

    I personally doubt that Google will discount EMDs in their results pages, for the very reason that you indicated – brand naming. If they fail to provide relevant results in their pages, people will use the other search engines that do show relevant results.

    It must be close to impossible to automatically sort out the genuine branding sites from the ones set up to benefit from EMDs, other than by human intervention. Further the EMDs generally do provide some valuable information regarding the subject (or should, at least), so Google will still be returning valuable search results for searchers of information.

    Cheers!

  22. Excellent post-Penguin summation Jonathan. Going to point to your key take away to a few clients who are in denial as to the severity of an anchor attack.

  23. The EMD’s may be standing up to this test but I doubt they will stand the test of time.

    Google are surely targeting EMD’s and with the amount of EMD “Made For Adsense” style sites there are it makes sense they would penalise them.

    However thanks very much for the research Jon. Truly appreciated.

  24. Excellent post! Great detailed analysis! It really does make sense that Google would use this method for ranking sites. I suppose non EMD will have a run for the money as time progresses.

  25. Jon,

    Excellent post and meticulous research. This share is very much appreciated, and gives some truly great actionable data.

    Keep it coming :)

  26. John, I find this contradictory to what I am experiencing right now. All my EMD’s with the exception of 1 has tanked, not even in the top 100. Now I realise that it could be the number of links, anchor text (EMA), velocity, link neighbourhood etc but right now I don’t think EMD’s have a great advantage.

    Also what happens in Penguin update v2,3,4 & 5 …

  27. Suggestion for another test to inner pages.
    Does the same conclusion hold true if the exact match is part of the URL (but not the top level domain). In other words, can I send a lot of backlinks with anchor text “retirement calculator” to the URL http://www.retirement-income.net/retirement-calcaulator.htm ?
    Thanks for the great work. Valuable for us all.

  28. Who’da ever guess that everyone’s biggest fear is muchado about nothing. This is in sync with what I’ve been hearing and your data solidifies this. Excellent post Jon, as usual. I always thought there was something odd about pandas and penguins.

  29. Hi John,

    appreciate the combination of your knowledge in this area and the research you have undertaken – this really adds to our comprehension of what is going on and cuts through the hear-say and BS.

    What you have found and the conclusions you draw in relation to Google’s activity certainly are logical.

    Most of my sites are very similar (Amazon affiliate) and EMD but there is some variability in how they have performed after penquin.

    Cheers and thanks again John,

    Mike FD

  30. Thanks for the interesting insights Jon.

    To me, the wide-ranging and largely irreconcilable responses to your article are equally enlightening. What seems to jump out is that given the many factors that affect search, it’s not possible to draw definitive conclusions based on data that isolates only a few of them.

  31. Thank you for this post Johnathan, it does throw some light on the situation, and I think that they are not punishing these sites because they are only using their name in the anchor text, and you can’t really be punished for using your name.

  32. Jonathan, thank you for your excellent article, work, and analysis. This is very helpful.

    I have been getting a few emails from people who have posted articles and links in my article and directory site stating Google will penalize them and me for having an article or link on my site! I’m flabbergasted and can’t believe the people are stating the sky is falling…well if Google talks…people listen as if the police were knocking.

  33. Good stuff but I wonder if the type of link (article directory, website directory, RSS, Web 2.0) has more to do with it than the anchor text.

  34. Hi
    I have over 100 sites and after a lot of testing i have found that all my wordpress sites that got hit bad google is no longer indexing any new posts to them.
    In the past a new post would be indexed with in hours or faster and none of them get indexed at all.

    I had many sites on page one and many excat match sites that all got moved to no place.

    My older sites i have not worked on in years have all moved up now and they will index a new post all wordpress sites.

    To me google seems to hate sites with a lot of backlinks now and a lot of new information in posts and is moving up sites that are junk with no new info and not much to offer.

  35. Great info Jon. Your detailed analysis was awesome as always.

    No doubt Google has been favoring EMDs all this while, but even with EMDs, I wouldn’t take the risk to have a too high % EMA.

    Have always diversified the anchor text for all our clients sites to look more natural. Better to be save than sorry. No doubt it’ll take more effort to rank for the desired main keywords, but the sites will keep ranking no matter what updates Google comes up with.

  36. I have several EMDs that fell off the map, and they used to be page one.

  37. Hi Jonathan,

    Just got to thank you for this detailed information for FREE.

    It’s incredible the amount of fluff that is being sold over at WF and based on pure speculation alone.

    Your a breath of fresh air Jonathan.

    cheers,
    Jerry

  38. Unfortunately my exact-domain site now ranks lower than non-exact sites… (it used to rank TOP1 before Penguin…)

  39. Hi Jonathan,

    Great post. I will also say that I have had a couple clients get hit with Penguin despite the EMD, and have managed to bring them back up to page 3 for some keywords.

    For those of us using software like Senuke X combined with TBS, how would recommend changing or refining the anchor text when building backlinks, or does this anchor text penalty only relate to on page?

  40. Thanks for the great info! Something else I have noticed is that Google does not like spammy backlinks. I went to fiverr.com and paid a gig to have 250 backlinks to one of my websites, which really pumped up the results from a “site:xxxxxx.com” inquiry. But it basically killed my site. It was indexed, but not ranking in this universe.

  41. Great analysis, Jon.

    When you said ‘Some Non-EMD Sites Are Also Getting Away With It’ I was wondering: Is it possible that they are using 301 redirects with non-anchor text backlinks, or would you have noticed that?

  42. Very interesting…

    Of few of my observations:
    The EMD’s and EMA’s need to have a consistent backlink plan as well. That is how they are ranking high, they have GREAT backlinks.

    If you check the backlinks on a few of those difficult KW’s/domains, you’ll find that their links are high Page Rank, high Page Authority, and have high PR & high PA sites linking to THEM (backlinks to backlinks).

    Couple that with EMD, and it’s hard to be ignored. Of course if you get nothing but low pr links in high quantity – it looks fishy and triggers the Matt Cutts spam filter from hell.

    Your thoughts on that Jon?

  43. Great post Jonathan. Thanks for sharing and I’ll be sure to take this info to heart when I’m looking for EMD names for new niche sites.

  44. I had a site get crushed. 6 keywords on page 1 to not in top 50. No warning about links in Webmaster tools either. I use SEOMoz and it got an A+ grade. It’s a totally legitimate business site. High EMA may have done it, but I SEO lots of other sites that had no impact and also had high EMA but not an EMD

    Penguin has me scratching my head. IMO, Google screwed the pooch on this update.

  45. Great info Jon. The first part of your post is over my head. However I get the second half about exact match domains. I have a few exact match domains that were ranking in the top two positions, but now they are at the bottom of page one after the Penguin update. They are static sites. The sites ranked higher than me now are all blogs, so it appears Google has lessened the weight of static sites. I will probably turn these sites into wordpress blogs soon and occasionally add fresh content.

  46. My top EMD with tons of EMA got knocked down from mid 1st page to #72 by Penguin.

  47. All of my EMD’s got hammered. They’re not even in the top 1000 search results. so they are off the grid (indexed, but off the grid).

    It’s nice to know that EMD’s did better than non-EMD’s, but it doesn’t change things for me. I have to start over.

  48. Hello Jonathan,
    Thank you for all of your great insights and clear and instructive articles.

    I agree with your statement above “In short, exact-match domain names are your friend!”
    I have a variety of small sites (3-5 pages) with exact EMD and EMA for the specific product that site is promoting and these have shown up on Google page one within 7-10 days from publishing.

    After Penguin some have dropped to #2 or even lower…
    I have very few back links since these sites are less than 4 months old.
    I am eperimenting!

    I look forward to your next article,

    Best regards,
    Peter

  49. Great post Jonathan. These findings match my own experience.

    Thank You!

  50. Hey Jonathan, thanks for this in-depth and very informative post. I am on your list and I have been reading your articles post penguin, and you really have some great stuff here. (I even sent out an email to my list recommending your post – though a very small list).

    I also read an article by Dori Friend and it seems to have some gems in there as well. She did some case studies and she has some shocking stats with regards to links from websites that are De-indexed. Don’t know if you are okay with this but here’s the link to her article: http://www.seonitro.com/easyblog/entry/pandapenguin-case-study-analysis

    If not, just edit out the links.

    Thanks for keeping us up to date and all the best.

    …..
    Casey Gentles

    • I wouldn’t ordinarily allow a link to a competing service’s blog, but it’s a great post that has some very interesting postulations, so I let it go this time. :)

    • Thank you Casey for that link and thank you Jonathan for letting him post it. all these case studies seem to boil down to a few simple rules. Diversify, diversify, diversify.. Don’t panic, diversify some more and always have your towel (Hitchhikers Guide) ~~ As you’ve posted several times, no need to back track, just pick up and move on. Social networking~~ SO BE IT! The less we have to dance around Googles latest song and dance, the better

  51. Excellent analysis backed up by data. It does make perfect sense when you think about the big brands and exact match domains not being penalized. Google has quite a dilemma with these last updates however. I think the general consensus is that we are not necessarily seeing the “best” sites when searching on Google. This fact compounded by negative seo issues and I think you have all of the criteria you need to roll-back the last algorithm changes. Of course that will never happen.

  52. Also curious if anyone has had success bringing a Penguin penalized site back to better rankings? I’ve had a few of my older, larger sites hit pretty hard and I’ve been de-optimizing the on page stuff so it is much more natural. I’m also diluting the backlink text but haven’t seen any improvements yet.

    I also have some newer, exact match domains that were largely unaffected and most have very few links to them. Most are largely in the top 5 for their main keywords now so I can concur with what you are saying here Jon.

  53. I think that EMD get a pass because it is natural for people to link to their site with a “naked link” example “www.mysite.com. The EMA would be a lot higher for these terms, so Google gives them a pass.

  54. Jonathan,

    Solid analysis. My experience is that Google quotas EMD’s per keyword. They select what they perceive to be the best and allow 1 EMD to rank in top 10 along with Wikipedia etc.

    Jeff

  55. I have an EMD that was ranking 1 and 2 for its term for 12 months. I was also ranking pretty good for a few other keywords and longtails. It’s now falling deeper and deeper into the abyss. My traffic is down from 200+ a day to less than 50 and my earning have gone to virtually nothing. Not that was making a living from it, it was around $300 a month. It was all natural backlinks and there was nothing spammy or blackhat about it.

  56. Interesting. I’m in the insurance niche and I’ve seen EMD’s ranking 1-3 in google with barely any linking profile at all and most are simply adsense sites with 3 or more blocks above the fold. EMD’s at least in the insurance niche even rank higher for their EMD keyword than PR 3-5 sites with a huge link footprint. Great info Johnathan as usual. Keep us updated if you drill down more and have more to share!

  57. Thanks so much for sharing this data Jonathan. It really shows what a powerful tool keyword canine is. Glad I picked it up.
    It would be interesting to see ( although tedious to check), what the quality of the content is of the sites that rank in your data. Call me naive, but I still believe google wants SERPS to lead to good content so as to best serve the searcher.

  58. Great info Jon, I believe EMD’s are getting a pass because if someone were to link to the site with a “naked” link, example “www.mysite.com”, the link would naturally have the EMA in it. So Google would have to give a pass to those sites in order not to trip a filter for what is considered a natural way many sites link to each other.

  59. Jonathon,

    My exact match domain (www.saveyourmarriage.ws) had enjoyed a middle of page one ranking for several months, and made it through the Panda update unscathed. However it fell off into the basement after Penguin. So it seems that in my case, the exact match domain did not get my site the pass that you wrote about. Bummer!

  60. Thanks for sharing, Jonathan. A quality post, as always.

    I was interested to hear you don’t use Google’s Webmaster Tools because you don’t trust them that much. I don’t either, but One needs to know how they view your site, whether they are unable to access pages, the keywords they believe you are attempting to rank for, etc. Now, there may be other ways of obtaining this information and I would love to hear your thoughts about this.

    Regards,
    Nick

  61. Simply brilliant! How long did it take you to do all that?

    • Probably four or five hours total time for the data gathering and analysis. But building the tools I used to do all of that, well, that took a lot longer. :D

      • Jon, this answer to the post above yours certainly makes a lot of sense to me. As it would be impossible for any of us to figure out how much time, it took you to build the tools! Keyword Canine alone was certainly a challenge, as were the many other tools you produce.

        Anyone who is an active member of your forum knows that it is constantly evolving, and that you are on/off of it many times a day making post that help ‘ALL’ of your members. From peon’s such as myself to ‘GURU’s’ in their fields.

        Working in the manner that you do, putting this much time into helping your numerous members it seems impossibility for you to find the time to develop all of the fantastic tools that you offer. To say nothing of the time you take to make all the informative blog post trying to keep us on the right track with ‘G’, along with all the newest information coming out on the internet!

        Your blog and forum post have helped me many times!

        HoneyJo

  62. If you want to be sure to get your pages well optimised and in good shape for Google, there’s a very simple solution for you guys and gals using WordPress ~ just install the plugin “WordPress SEO by Yoast”.

    Then inside you will find on every page a Linkdex Page Analyis tab which you can use to ensure you fully optimize your page correctly.

    Here’s some of the content from one I am working on at the moment…

    * The keyword doesn’t appear in the first paragraph of the copy, make sure the topic is clear immediately.
    * The keyword density is 0.98%, which is a bit low, the keyword was found 8 times.
    * The images on this page do not have alt tags containing your keyword / phrase.
    * No outbound links appear in this page, consider adding some as appropriate.
    * The copy scores 52.2 in the Flesch Reading Ease test, which is considered fairly difficult to read. Try to make shorter sentences to improve readability.
    * The page title contains keyword / phrase, but it does not appear at the beginning; try and move it to the beginning.
    * In the specified meta description, consider: How does it compare to the competition? Could it be made more appealing?
    * Keyword / keyphrase appears in 1 (out of 4) headings in the copy. While not a major ranking factor, this is beneficial.
    * The keyword / phrase appears in the URL for this page.
    * The meta description contains the primary keyword / phrase.
    * The page title is more than 40 characters and less than the recommended 70 character limit.
    * There are 845 words contained in the body copy, this is greater than the 300 word recommended minimum.

    Do yourself a favour and check it out.

  63. Being able to sort data to this degree on demand is enormously advantageous. Am curious, After keywordCanine mines the data what other tools are you using to sort the data or are you manually doing it?
    I agree with a previous commenter in that you are one of the few that, because of consistent value, reading through your entire post becomes second nature once your name pops up in my inbox.

    Many thanks, Jonathan.

  64. Analysis is great, but you can over analyze and never get anything done.

    Anyone trying to build a real business that will last should take their eyes off of what Google is doing and think about offering Value to their customers.

    Besides – Google search is getting less and less relevant as time goes on. If you offer lousy content in order to get people to click on the paid ads, well… that’s just a poor business model – and will never last.

    • It’s true that you can over analyze, and it’s truth that you shouldn’t focus on junk quality, but knowing what works and what doesn’t can help you if you’ve been hit by Google’s latest update. Google slammed a lot of very high quality sites in this last update (which is why their current results look so lousy), and many of those webmasters are working to restore their rankings.

  65. In the last little while, a lot of marketers were taking about what to do following the Penguin update, in accordance with their “testing”, without showing any numbers. You are the first marketer (as far as I know) who released some hard data on the subject. Congratulations and thank you.

  66. Interesting stuff. So I conclude that when distributing articles for link building purposes, it is important to spin the anchor text link as well as the content.

  67. Great article John. Lots of links with exact keywords WAS a great way to get to page 1 – but am I the only one that thinks Google has changed its corporate thinking now it has Shareholders to think about? Remember what corporate greed did in 2008, especially by the banks? Maybe Google is hoping to increase Adwords income through people who go from page 1 to oblivion. Or am I just a weird conspiracy theory idiot?

  68. Wow this post has so much information. But it has raised more questions but I now know that I can use expect match if my domain name is in it!!

  69. Very interesting, thanks. My main niche has an EMD that has hyphens and a “4-” in front of it. That has gone down from position 1 to position 3. Positions 1 & 2 are higher PR than me.

    Prompted by the comments in this discussion I have registered EMD.co and E-M-D.co.uk (I am in the UK!)

    Now, I’ve got them I’m not exactly sure what I’m supposed to do with them to get any benefit without mega-effort!! Possibly put a video on…. ? Single page that redirects to my main site, or to a relevant age within that site? Sorry this is a VERY steep learning curve for me :-)

  70. Thanks for sharing all this fantastic data Jonathan – I’ve noticed many people saying that EMDs don’t work anymore and this just goes to show that it is not true.

    I have an EMD that lost a lot of traffic after Penguin, but suddenly jumped back up in the results again a week ago (without me doing anything). I’ve still not worked out why!

  71. Hi james,
    I did my own research on 20 or so sites ranking well with EMA’s of 80-95%. Like you, I found some of these sites actually had very few backlinks and some had thousands with a very high EMA.
    One of my sites heavily penalised by Google has an exact match domain for very competitive keywords yet it still ranks on the first page for that exact combination of keywords but for every other combination of those keywords it’s outside the top 100 even though every combination ranked in the top five before.

    Having read your article, I’m still none the wiser as to why some rank well with 90% EMA and some don’t?

  72. Great sharing Jon, now we probably really need to look at exact match domain names. Thanks.

  73. Hi Jonathan,

    Great post, I’ve experimented with EMDs for a while, One tip I can share with you is that including a H1 title in your header with the same key phrase as is in your EMD seems to work too. Maybe that is down to the branding issue that your article spoke of?

    Steve

  74. Hi Jonathan,

    Must say you have done really good work for all of us here. Since the penguin update a lot of information has been going around. Your post has made a whole lot of things very clear for me. I’m in the middle of buying new domain name for my services page, this info is going to be a great help for me. Thanks for sharing.

  75. This is interesting, one question. What is EMA?

    Thanks for posting this, will pass on.

  76. Very very informative post. I have gathered many information from this post. Exact Match Domain has a great value in SERP. I have also noticed that. Did you check something with hypens(-) say seo service provider will show seoserviceprovider in SERP but not seo-service-provider, so I think (-) is not good for EMD. Again, thanks a lot.

  77. Nice post Jon.

    Have anyones “Penguin Penalised” sites come back?

    Jon – Have any of your penguinised sites returned to their former rankings yet?

    I am a little sceptical that simply by diluting your own sites backlink profile that sites will magically return.

    After all, the same tactics that could be used to quickly create exact match links, could just as easily be used to produce random, non-exact links.

    There must be more to it then that.

    There is the usual flurry of “Beat Penguin” WSO products etc. But I am yet to hear of anyone who has actually got their rankings bank yet.

    My own are still way down. Though I have noticed that despite the traffic staying low that the phrases that are ranking are hopping about a lot as though Google is playing about at the back end to twiddle with the results.

    At the moment the search engine results are awful.

    I even did a search on Bing the other day because they were so crap!

    They will have to tweak this quite a bit I think even for self-preservation reasons, because even average punters will start to notice after a while.

  78. Thanks for the article.

    I have an EMD that ranked on page 1 for “bed and breakfast Inverness”. The domain has a sitewide link on an PR5 site with a good few thousand pages using the anchor text “bed and breakfast Inverness”.

    Post Penguin the site no longer ranks at all for that term but the inner pages do rank for their keywords. It would seem that there is something more to how the penalisation is decided. My case may be a manual penalty.

    • Ness…I have found the exact thing to be true with one of my sites. The main page does not rank well at all after having been on page one for years, but interior pages still pop up on the first page of the SERPS. Strange. I am not sure what is causing this to happen, but it does.

      • The main page has been penalised but is still passing link juice to the inner pages

  79. Another great blog Jon… thanks for you ‘insights’ and well presented data.

  80. Interesting findings Jonathan. Logically this makes perfect sense, however I have an EMD that dropped out of the SERPs completely after Penguin for its main term. I have been busy trying to dilute the anchor text to get it back but to no avail so far. Like you say, I guess this depends on the niche. Fingers crossed that the dilution works and it comes back soon.

  81. Yes, we’ve seen this trend exactly across our network. The EMD has survived, even though the incoming link anchor in some cases is almost 100% non diversified. As you rightly say, if Google tuned down the incoming link text and the domain name, they’d be killing brands. Having said that, we’ve seen examples where Penguin has killed brands – mainly because the brand name itself has not been used enough in the anchor text.

  82. After post panda I cancelled all my blog network membership since big G slap almost all my site, it’s really horrible. With your analysis now I become clearer why my site never sit in page one again. Thanks for great content. I am waiting your content about social signal and how they affect all sites.

  83. Thanks Jon, I think everything is unfolding steadily. Somebody in a comment above called you a good detective he he he. I would say you have surpassed Sherlock Holmes himself. I see that you have done a great job and say thank you a lot.

    I was reading another post yesterday and I see that similar things have been up. One of my sites that was hit did not have any EMA in it and has a large number of % for the keyword I “was” ranking for. I think I will go ahead and reduce those where I can since I don’t have control over some places from where my links come.

    A well done job Jonathan!

  84. Jonathan,

    Thanks for doing all this work and sharing the results. Wrt diversifying your anchor text, do you have any data on what type of diversifying anchor works:
    - Closely related variants of the main KW targeted
    - LT KW based on the target
    - Brand related KW tightly related to the name or brand of the site
    - totally unrelated text (click here, check this out), naked url or even nonhyperlinked url

    Your targeted EMA for the niche might be 10%, but can we make the other 90% just closely related or variants of the main KW? This is probably more relevant for local/offline niches. You can probably come up with 10 variants of a local KW that are relevant & related (plumbing seattle, plumber seattle, plumber in seattle, plumber seattle wa, etc). Assuming this case, if all our anchors are just variants of the main kw and we link to 10 KWs at the ideal 10%, would that be just fine? Or do we have to allocate a significant portion to brand anchors and unrelated anchors?

    I’m just wondering here if you have some data about this. Perhaps this is also niche dependant?

    Thanks for sharing.

  85. Deny, get off this blog and go back to the day job. Great Post Jonathan. I have been following your information for a while, and posts like this show us just how much passion you have for this industry.

    The Google algo has changed, probably forever. However, many smart affiliates are still ranking like crazy and will always will find ways to do so. Smart minds never die. The game goes on….

    Best

    Johnny G

  86. Thanks for the great post Jonathan! I still can’t figure out the Penguin update and i have several EMD’s that have different outcomes after the update…so pretty confused here.
    Still your info makes alot of sense to me.

    Thanks.

  87. Don’t know if this helps but in an effort to “water down” EDA, I have built a blog network of some 30 blogs (on different C class ips) all on the same subject but using a plugin, every time the Google bot visits a page , it will find a different EDA than the main EDA I am trying to rank for. What do you think?

  88. Nice detective work Jonathan. Looks like I better snap up some EMD with the lesser domain extensions before the info here spreads far and wide – which I am sure it will.

  89. Great report Jon!

    Google use over 200 signals to rank websites in SERPs. For those over-optimized non-EMD sites that get away with penalty from Google, their links could have been built slowly and steadily from time to time.

    One question from me. If EMDs have those advantages as you mentioned above, would all the EMDs with different TLDs stuffing the first page of the SERP in the future?

  90. Good stuff. But did you think about correlating your data with on site factors. For example, http://www.plumbingsupply.com/ is clearly an authority site (2560 indexed pages). So the fact that it has not been penalized for 92% EMA may be because it has a ton of original unique frequently updated content.

    It may be that the multi-factorial formula takes all these in to account (on site factors and backlinks) and the final result is dependent on that?

  91. Great post, out of interest how much content did these sites have on them? See any particularly sparse?

  92. GREAT article. Thanks. I love analysis based on actual numbers. Keep it up.

    But you never mentioned if the number of links in the analysis is total or by domain. I expect it is total since you never specified. The thing is with this type of analysis that is based on percentages it is very important to first of all know what the numbers represent but also to compare the two, total links and links by domains.

    The percentages can be very misleading if for example a site has 300 links total and 75% EMA. But if 200 of these links comes from the same domain (a site wide sidebar link with EMA) Google will most definitely treat these links different than if they come from 200 different domains and all EMA.
    So the EMA % for links by domain is in my view also an important part to look at to try and understand this.

    Thanks again for a great analysis.

    • I said in the post that the number of links are the number of links from unique external domains — not duplicated links from the same domain.

  93. Very Interesting and Useful for the SEO guys. Thanks a lot for sharing the blog.

  94. Thanks for the great info.

    I not sure if this is true but some domain with age rank better than recent EMD.

  95. If you get EMD especially related to products name. Don’t setup your website with Autoblogging software, Google will de-index your website just like mine. Create manual posting, review style, divide to 4 section, products info, pros and cons, video and photo, snipet review from Amazon. You will rank top 3 on the first page, without backlink/s.

  96. Hi Jonathan, great insight.

    Well, I could say that 80% of my sites have been slammed. Few of them were glued to position one for over a year and now they are nowhere to be seen. They were at the top spot for the reason of genuine information, excellent content and interaction via comments. Some sites are bigger, some smaller but again this is against any logic I have experienced before.

    What do we do…..really?

    We go back and start editing good old trusted sites or simply wait for all this to blow over?

    Emma

  97. Hi Jonathan

    Great research, we need more of this in the SEO community.
    question: When you calculate EMA percentages, how do you handle if one domain backlinks from 50 pages with the same EMA? Do you count it as one domain linking with EMA or 50 links with EMA?

    Usually when I see very high EMA’s, it is due to a few domains with sitewide links, and Google might be more interested in the fraction of domains linking with EMA.

  98. Hi Jo

    Have you consider taking a look of those high EMA sites with less than 50 domains, and find out if their average EMA is a lot higher than the overall avg?

    I have a theory that maybe google have a threshold of number of links before applying the penalty. When you are just getting a few links it’s easy to have a much higher ratio of ema right? So if we can see that the avg ema is indeed higher if you have less than a certain number of link, we can sort of trust more about this theory.

    If it is easy for you, can try the threshold to be 30 50 100 and see if the result are more significant:)

    • Many of the sites have a large number of links (hundreds or thousands of unique linking domains), so it doesn’t appear to be based on some threshold.

  99. Excellent Post! Jonathan I must say your probably the only blog that I actually read all the words. Very good incite about the EMD’s. I’m of the Godaddy now. Bye!

  100. I love your methodical research you contribute to our SEO community. Thank you!

    Here is a site I can throw into the mix:

    As for the EMD/EMA findings, one of my “try out” sites http://www.toptenmostdangerousdogs.com which had about 18K traffic dropped down to about 1900 uniques/month does not seem to be in line with these findings.

    I had more than 100 keyword phrases on page one and two with about 30 on pos one, incl the EMA Keywords in the domain name. Post penguin, the kw’s rank much lower and are hardly found on the first two pages.

    I have about 400 bl’s from 200 root domains to the site, with 31 of them dangerous dogs in the anchortext.
    Anchor Text Total 61
    dangerous dogs 31
    http://www.toptenmostdangerousdogs.com 10
    devsdca 1
    gg 2
    2nd most dangerous breed of dog in the us 1
    2. 1
    the most dangerous dogs in the world 4
    2
    most dangerous dogs 1
    toptenmostdangerousdogs.com 1
    http://www.toptenmostdangerousdogs.com/ 1

    The site’s overall traffic tanked 90% despite a relatively low EMA percentage and for the three EMD keywords the rank dropped from pos 1/1/2 to respectively 41/50/57.

    Thoughts?

    GG

    • update: The BL data was from the Majestic historical index. the fresh index data shows 56 BL’s from 43 root domains…

  101. Denny…Jonathan’s here giving us his valuable advise, information and incite for free, and you get in his face about some petty lack of understanding on your part about a few word you don’t understand.

  102. One of my EMD’s got hit hard I’ve tried to bring it back with no luck as of yet. The site sitting at #1 for this keyword (same guy has 4 sites on the 1st page talk about an over achiever lol) anyhow I did a small test just to see if I could affect the number 1 site. Given that he has 4 sites in the first 6 spots I figured it would just move his other site up if the #1 dropped off. I hit his site with 50k blog has not the been the least bit affected however mine which is an EMD is no where to be found karma? I think so lol.

    Thanks for this insight Jonathan. I’ve been trying like mad to read everything I can post penguin and your posts seem to always be spot on!

  103. Jonathan great analysis however I think there is one thing missing. I believe if you are over optimised on anchor text it’s fine if you have a low volume of links. This would give another explanation for some of the EMD results you are seeing.

    This HAS to exist. Imagine if you got 5 ema links from a total 5 links and you then get punished, just doesn’t make sense. I think you have to have x amount of links before it kicks in.

    You may want to links to Dori Friends post as well which came out yesterday an was excellent. I believe you got a mention.

    • Some of the EMDs have hundreds or a thousand or more domains linking to them Joe. For instance, workfromhomeindia.biz has 202 linking domains, musicmanagement.com has 394, webdesigncompany.net has 1,560, etc.

  104. Great stuff again Jonathan! Very interesting regarding the EMD, and makes a little sense in my current experiences. I have two ‘money’ sites within one niche. One is 18 months old, reasonably newer than other sites but roughly double the site size. It is a partial EMD, with one of the 2 word keyword combination, plus a descriptor. Sits solidly at position 5 P1 post penguin.

    Here is the interesting bit, my second site is EMD123.co.uk, 7 months old, not many links and around a third of the site content of my first site. It sat around position 14 pre penguin, and moved up as high as p10 post penguin. However, 2 EMD have had sites launched on them post penguin, both domains are older than my 123 domain, one has very few links, the other has had quite a few built.

    The effect? My EMD123 has slipped to P22, whilst the other two EMD have appeared on Page 1.

    Do we think Big G is actually comparing EMD against each other specifically or am I being paranoid?

  105. I am starting the process of link diversification for some sites, and removing links for other sites where it relates to penguin. I am removing links that would not pass manual checks for sites that have received webmaster tools notices.

    However, for one site, I removed the links several months ago, and webmaster tools still records them as being links – even when sites have been decached and domains let expire.
    - How long does it take for Google to remove the links from webmaster tools?
    - If I do a reinclusion request while the sites are still mentioned on there, am I running the risk that Google will see some of the remaining live article sites and decache those sites while it is at it? Would not be the nicest.

    • I don’t use Google Webmaster Tools, and I certainly never communicate with Google about anything relating to my sites. I don’t trust them further than I can throw them.

      • Jonathan…

        So, I assume you don’t use google analytics? I only ask because some people like to argue that your stats or numbers are not legit unless they come from google analytics and I don’t agree. Flippa is a good example of where this debate happens at times.

      • I get not using GA ..

        But I dont get the problem using WT .. All the data in there is from Google to you, compared to GA, which is from your site to Google.

        If anything WT allows a webmaster to spy a little on what Google is doing, and it’s only of the only places to get historical rank changes and see different pages ranking for terms and placement.

        Just not sure how that can hurt you – other than if you tie all your accounts to the same account .. That may not be a great idea :)

  106. Heh jonathan another quality piece of info!! thanks again for the well researched info

    quick qs for you on emd? a while ago i was given the advice that if you cant get an exact match domain what you could do was add an extension on the end ie: add an x or a v as google wont read the x or v

    ie: http://www.workfromhomev.com

    can you advise on that please

    thanks again
    stephen

  107. you may missed on a crucial element of ranking: website age, i saw some of those sites even made before 2000, so they ranking for sure even with little backlinks no matter EMA or not,

    • Perhaps, but many aren’t that old. For instance, plumbinghowto.biz (2012), homeimprovementsmaryland.net (2011), workfromhomeonline.us (2009), etc.

      • I have made the same conclusion about EMD:s and bought an .biz EMD after penguin and the site is already on page 1 for an 30 000 exact match search volume/month keyword. I have only 5 links pointing to it. EMD:S work awesome now.

  108. Great article. I reported this issue on your forum site too. My site is not EMA for primary KW and got droppet form #1 til #11 overnight. This is a quality site. That do you suggest? Moving to new domain and do a 301 redirect to new domain?

  109. Well done on the comprehensive analysis, however my experience doesn’t correlate.
    I have a number of EMDs and all have suffered to varying degrees – none of them maintained or improved their rank.
    Maybe I am the exception that proves the rule?

    Example:
    42inchlcd.net ranked on page 1 for “42 inch lcd” up until late April.
    Now it has disappeared completely (not on first 10 pages), although the site is still indexed.

  110. Very nice analysis. Right now exact match domains are doing fine. But somehow I have the gut feeling that next time Google updates (maybe Zebra-update – another black and white animal) they might target all the crappy exact match domains that don’t have authority. Right now there are way too many of them out there and 99% of them are crap.

  111. Very interesting blog post backed with data.

    Better than some of the paid junk out there.

    I do have some exact match domain named websites that tanked after the updates probably because I ‘over optimized by too much I guess.

    Based on what Jonathan shared I am going to try to see if I could ‘neutralize’ my over optimization and sees if I could get my exact match domains back on track.

    Thanks

  112. Interesting findings Jon and yeah makes perfect sense as you pointed out. Interesting that other domains other than the big three are also ranking. It would also be interesting to see a comparison between anchor text that has traffic/value vs anchor text which doesn’t such as click here, see this post, etc. Or even URL vs anchor text.

  113. Hi Jonathan,

    Thank you for presenting your original research. We too did some analysis on a specific set of keywords and came up with similar results.

    One difference we did find compared to your work is regarding exact match domains on “lesser” TLD’s. Our researxh shows that these have taken a bit of a hit in the latest update.

    It’s a bit on the technical side because we need to be sure what we are saying is factual but here is the link which may be of interest – http://www.sayulabs.co.uk/how-have-the-recent-google-panda-and-penguin-updates-affected-rankings/

  114. Great post Jonathan thanks for the research.

    I am interested to hear your thoughts on the effect that panda and penguin have had on the use of spun articles. Do you think that spun articles are being recognised as such by Google and consequently hurting the websites that they point to as opposed to helping page rank.

    • I have not found that well spun articles are being devalued no. That only makes sense because they read like any other article does. I use them in my own backlinking.

      Trash spun articles I don’t use, so I can’t speak with authority about that. I can say that I’m still seeing sites that use such garbage ranking well in many cases, though. I don’t recommend it, however.

  115. Sooooo helpful – thank you! (And good news, since I have quite a few exact match domains – that incidentally are all still ranking pretty well.)

  116. Hi Jon,

    Thank you so much for the great post.

    However, I’ve got one EMD site still got penalized for the exact match keyword and variations. Can I link the site with many “click here” to recover it?

    And btw, is your keyword canine service able to sort out the average anchor text percent for each market?

    Thanks!

    Henry

    • Mind sharing the domain name so I can look at it’s link profile?

      As to recovery, diversifying your link profile is definitely important.

      And Keyword Canine does show the anchor text percentage for each URL ranking for any given set of keywords, yes.

      • Sure it’s greenlippedmussels.org and it ranked well for green lipped mussels and other variations. I used to throw a lot of BMR links on the site. After the whole network got deindexed, the rankings dropped a little, then the Penguin update and all my rankings are lost. Will Google penalize sites that was using BMR so that they will never be recovered?

        Thanks!

        Henry

  117. I would like to know if you noticed whether adding a “x” or a “v” to the end of the exact match domain name also works (when the exact match domains you desire have already been registered” e.g. workfromhomex.com when workfromhome.com is not available to register

    • There are a few examples of that in the data. For instance, business-management-software-guide.com is ranking for “business management software” despite having a 65% EMA.

  118. If EMA is so important, why don’t you have the sense to at least tell your readers what EMA stands for? Should we guess that it means “exact match authority”? I am so sick of reading something that is abbreviated without the author going out of his way to “communicate”.

    • Read the post. :) It’s in the paragraph just above the first heading:

      “(For the purposes of this blog post, I’ll refer to the percentage of links whose anchor text exactly match the keywords the site is ranking for as its EMA. Also, all of the queries were done via Google.com, so the ranking sites generally favor the USA. Lastly, all of the linking data was gathered using KeywordCanine.com.)”

      For you guys who like to scan instead of read, I’ve made the text larger and bolded the first occurrence of “EMA”.

      • I must say it is a marketing lesson in itself how you can gleefully reply to a high-minded non-reader like Denny.

        • I forgot to say thanks for your research (again) & debunking a couple of other previous SEO no-no’s I had been led to believe, eg. using hyphens and the viability of the lesser domains.

    • Bad day, Denny? Jonathan put a lot of research time into this post, which, by the way, is very helpful. If you just want to skim over something like this, then just leave as the point is to get into the details so you can become better at staying ahead of the curve.

    • Goodness!

    • Dude!

      Are you really going to dog on Jonathan for not holding your hand and using your brain for you?

      He’s spent all this time and tons of his resources to give everyone this data. He could have just kept the data to himself and used it for his own business.

      But he’s giving it away on his blog.

      Not only that, but he’s giving his analysis of the data. Analysis which is based on him running multiple online businesses. Heck, I’ve given him a couple hundred bucks this year already for software.

      He was talking about anchor text the whole time throughout this post. Then he uses an abbreviation and you get your panties in a wad?

      You come to the conclusion that “A” means “authority”… huh?

      Tell me again why you’re “so sick of reading something that is abbreviated without the author going out of his way to communicate”.

    • If EMA is so important, to you, why don’t you have the sense to at least read the article so that you’ll know what EMA stands for? Should we guess that you’re an idiot or just accept the evidence of our own eyes? I am so sick of reading something written without the author going out of his way to read what was written ~ i.e. your comment.

  119. Great post johnathan. I was starting to thinking about it because i have a EMD and it didn’t get the penalty.

    He is ranking 5. Two weeks ago it was 8.

  120. EMD, vey nice!
    But who knows what google is going to do for the next month.
    Now I am diversifying my traffic via web 2.0 property.
    Don’t put all eggs on one basket.
    Thanks for your article.

  121. So does it mean that we can penalize any of our competitor’s site by building links of same keywords?

    How Google can believe that all the links with same keywords are created by that site or by competitor?

    • Matt Cutts has stated that it is very difficult and rare for so-called “negative SEO” to work. He said that a site with otherwise positive signals won’t be affected by such attempts.

      I don’t believe everything Matt Cutts says, to be sure, but it only makes sense that Google would prevent that kind of thing from happening — otherwise it would be World War III going on between webmasters trying to knock each other down.

      Here’s a link for reference (though it has a lot of PR baloney in it too):

      http://searchengineland.com/google-talks-penguin-update-recover-negative-seo-120463

      • I’m not sure how they can prevent it. I am by no means advocating the use of negative SEO in your business, but let’s be honest.

        Last month, a large number of sites lost page one rankings and traffic as a result. This was because of off-site factors. Matt Cutts said this himself.

        Tell me how it is that Google can determine who is linking to their own site and who is linking to a competitor to kick them out of spot #1?

        I’m sure that a site starting with a 1,000 links (on page one) and a fairly normal anchor text profile would suffer from 4,000 links with EMA built over time.

        The deeper issue to the above situation is the battle that Google is losing with their algo.

        I truly think that they release these updates in pairs to make it confusing to figure out ranking factors. Imagine if people could manipulate anchor text and also build links quickly to a site – as a result they would get exposure to your users. Due to the low investment amount to get exposure, the sites were fairly low quality.

        Sound familiar? Wouldn’t spam quickly enter the marketplace? Isn’t this what happens when web exposure is as easy as A, B, C?

        Instead of tightening the rope on just one variable in your algorithm, you do it to multiple things. Now you’ve got people coming out of the woodwork, scratching their heads, asking what the heck is going on. One friend tells you it’s this. Another guys says it’s something else. Throw in blog networks being targeted and you’ve got yourself your intended confusion in the industry. Time for you to keep tinkering away to figure out a better search engine that will produce more useful results.

        If marketers start figuring out a way to manipulate your algorithm, which they will, you correct it.

      • Hi Jon,
        I don’t normally comment, but this article is pure gold, so I gotta make a bit of contribution.

        With some unfortunate Penguin side effects, Google may well have inadvertently created a whole new industry – Negative SEO.

        Here is an article of Successful Negative SEO Case, written in a fine detail by hired gunslinger, who was hired by a company to destroy their competitor. Quite disturbing really, but at least we better know how it is done and possibly be able to defend if somebody is after us or our clients

        http://trafficplanet.com/topic/2372-successful-negative-seo-case-study/

        I found also a very useful article “How to Rank After Google Penguin …” by Chris Rempel – the guy recommends listening Jon Leger’s advice, so he must be OK :) .

        http://www.thelazymarketer.com/blog/2012/05/09/how-to-rank-after-google-penguin-and-how-to-recover-if-your-site-got-hit/#.T7tOI8X4J7g

        Finally, Chris gives a link to Matt Cutts’ answer to:
        “How do you protect against sites being penalized for bad links that they have not built?”
        Matt gives a rather convoluted answer and maybe you’ll come to a different conclusion, but but I think he is saying: “we don’t give a &%$#”.

        http://www.thelazymarketer.com/blog/2012/04/09/google-directly-confirms-negative-seo-is-possible/#.T74KD8X4J7g

        Best regards
        Alan

    • Yes, you can bet that negative SEO is now being done!

  122. If a site was hurt with penalty, what would be the best way to regain its positions?
    Should you wait? should you deoptimize link anchor with branded links?
    Another option maybe?

    Thanks

    • It’s guilt by association – a penalty that I always doubted because it didn’t seem logical to me. As to answering your question… I honestly can’t say what will regain rankings after the penalty is applied. But here is my take on Google’s reasoning…

      After multiple requests to Google about the penalty, along with tons of other webmasters, I received the notice about unnatural linking. The notice went as far as telling me to have the webmasters who were linking to me, remove the links before being reconsidered.

      So there you go. What’s to say that my competitor didn’t have it out for me. What’s to say that they wanted the top spot so bad, that they linked to me over and over and over again. Maybe even linked to me for months with blog network links. Then one day, Google starts deindexing sites in the blog network. They decide that sites being linked to, must have been participating in the “scheme” to manipulate their index. Bam! Push them down and slap them on the wrist. Don’t deindex them because their sites might be useful. However, let the webmaster know that there is a penalty for manipulating your business.

      Most of the lot will be businesses who were actively participating in the “scheme” or paying someone who was doing it for them. Let the small percentage of businesses who lost rankings, file complaints, send out scary letters to the sites linking to them, etc…

      The fix might be to organize and get everyone to take down their deindexed blog network blogs. Or we could 301 redirect them somewhere just for kicks.

    • I have had a EMD that has been penalized. It was #1 before and I am struggling to get it back. It has lots of EMA links.

      I have another EMD that has gone to #2 and has much less EMA.

      What are the suggestions to get a EMD back?

      Thank you for sharing you research.


Trackbacks are disabled.