Internet Marketing

Do exact match anchor text links help or hurt?

  • Steve P Reply

    Obviously there has been a lot of chat about EMAT since the Penguin/Panda invasions. I think today's article on SEOMoz on the subject is pretty interesting: http://www.seomoz.org/blog/postpenguin-anchor-text-case-study

    Of course it's hard to draw any conclusions from a single sample, but wasn't surprised at the outcome. My question here is what do you think? The author's conclusions seem to be in stark contrast to Hubshout linking strategy (again, I'm not saying either is right or wrong - just looking to start a discussion on it).

    I reviewed 5 submissions at random on an SEO account with Hubshout and found that each article has 1,2 or 3 links in it. 1 link is always an EMAT pointed to the target site, and any remaining links are EMAT pointing to an authority site in the general "business" category.

    Whereas the SEOMoz author concentrated on almost zero EMAT links on related sites, the links I'm seeing are all EMAT on sites with little relevance to the target stie (the last 5 posts were on Alexandria plumbing, dentist reviews, veterinarian website design, Houston drug testing and New Jersey dentist)

    What are your thoughts? Has anyone else been able to test the effects of exact match anchor text vs "random" anchor text on relevant sites?

    01/16/13 at 01:27 PM (4819)
  • Adam Reply

    [quote]1 link is always an EMAT pointed to the target site, and any remaining links are EMAT pointing to an authority site in the general "business" category.[/quote]

    Not exactly Steve... We let the client / reseller select the anchor text ratio for this reason. What you are saying would only be true if "Use Keywords as Anchor" was set to 100% (which is not our default or recommendation).

    Also, the only time you have co-citations to a general category site is if you don't have specific co-citation requirements or writer research completed. I'd say that over 80% of our SEO campaigns don't use the general category co-citations and have specific plans built for them that are targeted to their industry...

    On your last point - We have a strong portfolio of category-specific media sites that carry rankings themselves. We've benchmarked our competitors - and they don't do ANY category-specific stuff at all. We do also syndicate to generic blogs and article sites as well - since clients generally want volume. But our emphasis is on finding industry-specific placements for clients.

    Conclusions on Anchor Text For every article I've read that says don't do exact match, I've read several that say do. In the end, we leave this up to our customers but recommend being in the middle somewhere to maximize diversity of backlinks and anchor text.

    Don't you love the SEO business? LOL

    01/16/13 at 02:01 PM (4821)
  • Steve P Reply

    Hey Adam,

    Yeah, you gotta love SEO. You are right in that fact that you can find support for almost any opinion out there if you search hard enough ;-)

    I disagree with you regarding the EMAT links. Our "Keyword Anchor Text Ratio" is set at 75% (I never even knew that existed until just now). But a random sampling of submissions in January are all 100% EMAT. I'd prefer that no more than 50% of links to the target site are EMAT - currently it's 100.

    I would like to hear more about our co-citation options and would be curious to know if any industry specific placements for the site in question exist.

    Anyway - I don't want to further derail this into a conversation about my site specifically. I'd like to hear if there are any other opinions regarding the use of EMAT links vs non-EMAT (yeah...that's a word ;)

    01/16/13 at 02:13 PM (4822)
  • Adam Reply

    Cool.. Let's definitely drop your campaign to 50% (Terrance can help you). Remember that this is an average. So just like Heads vs Tails - any one string will come out slightly off.. If you were at 75%, 100% on a sample of 5 or 6 is quite likely... As far as I know, the feature is working wonderfully.

    Terrance can also talk to you about WRITER RESEARCH and CO-CITATIONS that are specific to the campaign. It's one of my favorite features from 2012... It is REALLY AWESOME. We've had HUGELY positive feedback because it makes the content marketing so much better and useful for the end-reader.

    01/16/13 at 02:22 PM (4824)
  • Terrance Reply

    I have done this for you Steve. In my conversation with your employee she said that she would like to see them at a 100 also I would love to talk to you about our Co-Citations. Again, when speaking with "C" I mentioned it but didn't get much feedback on it.

    Thanks for sharing the article Steve. Very insightful read.

    01/17/13 at 09:37 AM (4838)
  • Adam Reply

    Steve -

    You might be also be interested in some of my recent experiments with Exact Match Domains. While a slightly different topic than Exact Match Anchor, this is also something that Google has publicly stated carries LESS weight in their algo based on a change last year.

    My experimentation shows that either Google completely misled us, or has undone their Exact Match Domain release from last year as they are shooting up the rankings faster than ever. My theory here is that perhaps they are de-valuing anchor text (which they have hinted at) but INCREASING the value of Domain Name and onsite words to compensate (which they have NOT said publicly)... After all, they have to keep their search relevant or they will suffer losses in users... There are only so many ways to figure out what a domain is about and how popular it really is! :-)

    01/17/13 at 09:45 AM (4840)
  • Steve P Reply

    My personal theory is that exact match anchor text and exact match domains will decrease in significance this year. Those are both just too easy to manipulate. More influence will be placed on links (and mentions - not linked) from topically relevant, high authority sites no matter the link text. Also, social signals should see an increase in influence, but only those from relevant sources or high profile, trusted individuals and entities.

    IOW - the trend of quality over quantity that can be spammed will continue. (duh)

    Anyway, I'll put down my crystal ball now and get back to work :)

    Cheers!

    01/17/13 at 01:34 PM (4858)
  • Chad Reply

    Adam and I recorded a video on this topic today. Steve, I don't disagree with your predictions but I'm always surprised at how slowly the ranking factors change. I think there are only so many predictive signals and Google turns them up and down in their quest for the best search results. That doesn't mean you should ignore the trends but focusing on a future state that doesn't exist could hurt you in the short run. There are always trade-offs. With that... here is our video:




    01/18/13 at 12:02 AM (4888)
  • Jenna Reply

    What a great topic.

    01/18/13 at 10:55 AM (4890)
  • Steve P Reply

    Hey Chad,

    I totally agree regarding taking into account what works now vs what may work in the future. Sometimes it amazes me still at how easily manipulated rankings can be. Of course the Google engineers probably don't see it as such as having such an easy solution :)

    01/18/13 at 11:43 AM (4893)