We usually do not comment on the algorithms of other search engines, but have made an exception today as Google has improved the way in which it interprets anchor text within a link.

Last week, Google announced a series of algorithm changes, through their blog post: “Newest search quality updates”. Within the 50 updates they have shown, there are all sorts of changes, such as better scoring of news groupings and improving the user interface on mobile. However, it is the updates to the Anchor Text that really stand out for us.

You might be forgiven – on first read through – for thinking that Google had decided to drop anchor text as a ranking factor when it writes:

Tweaks to handling of anchor text. [launch codename “PC”] This month we turned off a classifier related to anchor text (the visible text appearing in links). Our experimental data suggested that other methods of anchor processing had greater success, so turning off this component made our scoring cleaner and more robust.

Google’s wording is not exactly transparent here. They have NOT said they have switched off anchor text. They said they turned off a “classifier”. Without having a full list of how Google classifies anchor text, this is quite hard to interpret. Classifiers might relate to the location of the link on the page, whether the anchor text was there when the page was created (as opposed to being added at a later date), whether the anchor text is simply a duplicate of the Alt text or all sorts of other factors. In fact – pretty much the ONLY thing that we can deduce from this paragraph is that Google DOES still take links – and anchor text – seriously.  Google have simply changed the way they score anchor text processing.

The second change that they listed in the post about anchor text makes this even more clear – even though the paragraph is shorter and less instructive:

Better interpretation and use of anchor text. We’ve improved systems we use to interpret and use anchor text, and determine how relevant a given anchor might be for a given query and website.

My instinct is that these two paragraphs refer to the same algorithm change, but did not get de-duped before publication (is there a penalty for that?) but either way, this second paragraph clearly alludes to the fact that Google have IMPROVED (not discontinued) their anchor text evaluation systems. Clearly – however you look at it – Google does not treat all links equally and any tools you have to sort the wheat from the chaff will be invaluable for SEOs moving forward.


Dixon Jones
Latest posts by Dixon Jones (see all)


  • SEO Services

    hi, thank u so much for this info.

    April 10, 2012 at 12:06 pm
  • Douglas W. P.

    this post is circling the world – Brazil

    April 10, 2012 at 2:03 pm
    • Dixon

      Yes – it’s getting attention. Surprised nobody has offered ideas as to what Google’s “Classifiers” might be. given that they called the update code name “PC” I wonder if that stood for “Position Classifier” or “Post Content” to signal the link was added later or one of a hundred other thoughts that spring to mind. I guess – though – that over analyzing something like this would probably not be a good use of time – given that Google also gave 48 other small changes in March alone to their site and algo.

      April 10, 2012 at 2:10 pm
  • Chris Reilly

    If I were to speculate what anchor text classifier may have been modified I could guess it would pertain to its context… I read *somewhere* that text that surrounds an anchor may be used as a factor to determine its relevance. If something was rolled back, this would make sense to be it since there are so many different scenarios that it would be difficult to engineer a uniform way to interpret context. Just my $.02.

    The sites we promoted with cheesy exact match anchor text links still rank, so I imagine that as usual the SEO community is overreacting.

    April 10, 2012 at 4:51 pm
  • Miguel Salcido

    Those ARE two distinct notes on anchor text and not simple an “error” on Google’s part. The first one relates to the usage of the hyper link “title” that many people add to text links. This is basically an alt tag for text links and people have been keyword stuffing these for eons.

    The second one I believe refers to either the surrounding text around a link and/or some sort of measurement of the anchor text and the relevancy of the page being linked to.

    April 10, 2012 at 4:51 pm
    • Dixon

      You sound pretty sure on the first one, Miguel. Did you test that first hand? (I always find it useful to add “could be” and “May be” whenever I suggest anything connected to Google – but what you say sounds plausible.)

      April 10, 2012 at 6:35 pm
    • Ayana Meyer

      >”The first one relates to the usage of the hyper link “title” that many people add to text links. This is basically an alt tag for text links and people have been keyword stuffing these for eons.”

      –> sorry Miguel, but it´s not about the title, they were very clear saying ” related to anchor text (the visible text appearing in links).”

      April 11, 2012 at 2:47 pm
  • Jonathan Bentz

    To point no. 1, it could be that Google adjusted the overall value of keyword rich anchor text effecting the ability of a site to rank for said target keyword phrase. Some keyword groups that I monitor have seen big brands “pop back” into the ranks with limited anchor text links but many quality links – so it makes sense in that case.

    If Miguel is right about point no. 2, this would help weed out low quality blog networks that place English words in a 250-word post and then place anchor texted links at the footer of the post… so it would make sense.

    April 10, 2012 at 9:30 pm
  • David

    I’ve got to say I agree with the 2nd one referring to the content surrounding the anchor text. That would be a simple way for Google to detect and devalue any anchor text links surrounded by “spamglish” articles and blog posts.

    April 11, 2012 at 9:49 am
  • Jonny Ross

    Does google really clarify anything! I think this refers to ensuring you include brand anchor text, long tail anchor text, and random anchor text i.e. not the same anchor text to the post every single time.

    this for me is a common sense one, and google are going to be removing the black hat spammers!


    April 11, 2012 at 11:51 am
  • Justin Butcher

    This is pretty much the same point as the one I made here:


    April 11, 2012 at 2:23 pm
  • Phill Ohren

    On the brighter side, at least they let us know.

    Maybe it was a vein attempt to discourage link building?

    IMHO, I believe this change is aimed at the freshness of links. Google has made heaps of updates in recent year that focus on time-based metrics. Theres no reason why links should be counted or valued over time either. Still… it’s a mystery.

    I guess we should do some testing….

    April 12, 2012 at 1:18 am
  • Harry

    Im just asking myself why the hell so much SEOs over analyzing this little change? Just place a few Links in the time you ruminate over this change…

    April 13, 2012 at 6:48 am
  • Jeff Zelaya

    At the end of the day how will this impact my website? Will this change your SEO practices? How about your linking campaigns? If this will significantly impact you then maybe it’s time to reassess your whole SEO campaign anyways.

    April 16, 2012 at 9:36 pm
  • Vacante 2012

    At least Google is telling that they changed something, even though we didn’t have the complete picture of how that something worked before the change. On the other hand, that’s how seo works, it’s not an exact science and it should stay this way.

    April 18, 2012 at 8:53 am
  • Heather Buckley

    I remember reading recently that Google is looking for anchor text within body text that is not natural. I think varying anchor text will become even more important and that using natural language in the body text as anchors to RELEVANT content will become the most powerful way to build links.

    How will that impact your present search results? If you have used content marketing to add links that are not particularly relevant to the subject of the post and have repeated this using the same anchor text across many sites, those links will now be ignored or demoted.

    Best way to build links in my opinion is to build relationships with trusted and relevant sites and use varied anchor text that describes the content that the link is pointing to. Make sure the link is relevant to the article or post in question.

    April 26, 2012 at 12:02 pm
  • Peter Widmer

    Do you have news about the anchor text? What was the impact in the past? Any suggestions how to handle that now? Thank you for more information about your experience.

    September 17, 2012 at 5:00 pm
  • Marketing Tampa

    If anyone has questions as to the effectiveness of anchor text and Google rankings, they only need to look at a single Wikipedia entry. They probably do anchor text better than anyone, and it pays off big in the search engine results for them. I think that what they do with their internal linking and anchor text has a bigger impact than new content being added frequently.

    Google seems to do well with the anchor text I use, however Bing and Yahoo take more time identifying what I’m trying to tell it.

    November 17, 2012 at 4:06 pm
  • Sean Vandenberg

    Thanks for sharing this. I really like your comment on going back and hyperlinking text AFTER publishing the content. That gets me thinking…

    December 1, 2012 at 1:00 am
  • infatex

    Thanks for such clear explanations!Duplicating page title exactly is not essential. The important thing is that the anchor text contain keywords, however it is worded. Hope that helps!

    December 18, 2012 at 5:10 am

Comments are closed.

If you have any questions in the meantime, please contact help@majestic.com
You have successfully registered for a Majestic Demo. A Customer Advisor will contact you shortly to schedule a suitable time to connect.