A Data Driven Guide To Anchor Text (And Its Impact On SEO)

David McSweeney
David is the blog editor here at Ahrefs, the owner of Top5SEO and a white hat SEO evangelist. SEO case studies make him a lot happier than they should, and he has a tendency to overuse ellipses...

    There’s a lot of information out there about anchor text.

    But depending on which guide you read, you might think that you should:

    a) Build lots of anchor text rich links to shoot your site to the top of Google!
    b) Avoid building any anchor text links or you’ll be banned forever from search and put on an FBI watch list!

    So which is true?

    Well, like many things in SEO, the truth lies somewhere in between. So today we’re going to deep dive into the topic and, as usual, rely on data (rather than opinion) to form our conclusions.

    Let’s start by defining exactly what anchor text is and how it influences search rankings.

    #

    Anchor text refers to the visible, clickable words used to link one web page to another.

    anchor-text

    Example: In this sentence, the blue words are the anchor text.

    #

    Let’s say Ahrefs published an article that we wanted to rank for the keyword “marketing toolset”. Here are some of the ways that other websites might choose to link to us:

    Exact Match — the anchor text is the exact keyword phrase we are looking to rank our article for:

    Ahrefs is a data-driven marketing toolset powered by a huge index of backlinks, keywords and content.

    Partial Match/Phrase Match - the anchor text contains the keyword phrase we are looking to rank for:

    Ahrefs is a data-driven marketing toolset powered by a huge index of backlinks, keywords and content.

    Branded — the anchor text is the name of our brand:

    Ahrefs is a data-driven marketing toolset powered by a huge index of backlinks, keywords and content.

    Naked URLs — the anchor text is a ‘naked’ URL (i.e. as it would appear in a browser):

    Ahrefs (https://ahrefs.com) is a data-driven marketing toolset powered by a huge index of backlinks, keywords and content.

    Generic — the anchor text is an unspecific, generic phrase which does not include our target keyword (i.e. ‘click here’, ‘this site’, ‘this article’)

    Ahrefs is a data-driven marketing toolset powered by a huge index of backlinks, keywords and content. Click here to go to their site.

    LSI Keywords — the anchor text is a synonym for, or closely related to our target keyword.

    Ahrefs is a data-driven promotional toolset powered by a huge index of backlinks, keywords and content.

    Other types of anchors include:

    • Author — your name would be the anchor.
    • Title Tag — your page title is the anchor. Often these will also be partial match.
    • Image links — the alt text may be counted as the anchor.

    By running the “Anchors” report in Ahrefs’ Site Explorer, we can see that several of the anchor text types listed above have been used to link to our SEO tips article.

    Site Explorer > Enter URL > Explore > Backlink profile > Referring domains > Anchors

    anchors-report

    Now let’s take a look at how that choice of anchor text may influence our rankings.

    #

    Search engines (from here on I will refer to Google) use the anchor texts linking to a page as an indicator of the page’s topic; i.e which keywords it should rank for.

    In simple terms, if I was to link to a page from this article with the anchor text “dog biscuits”, this would indicate to Google that I (or Ahrefs) believe that the linked site is a good fit for that query.

    And again, to simplify:

    The more sites that link with the anchor containing the phrase “dog biscuits” (exact or partial), the more certain Google will be that the linked page should rank for that query.

    But of course, nothing in SEO is ever simple. At least not any more.

    Early Google’s (Over) Reliance On Anchor Text

    Anchor text was weighted heavily in Google’s original algorithm.

    In their 2008 paper “The Anatomy of a Large-Scale Hypertextual Web Search Engine”, Google founders Sergey Brin and Larry Page explained:

    The text of links is treated in a special way in our search engine. Most search engines associate the text of a link with the page that the link is on. In addition, we associate it with the page the link points to. This has several advantages. First, anchors often provide more accurate descriptions of web pages than the pages themselves.

    (yes, they were much more transparent back then)

    Using anchor text would also allow Google to determine the topic of (and return in search results) media formats where on-page signals could not be used.

    Second, anchors may exist for documents which cannot be indexed by a text-based search engine, such as images, programs, and databases. This makes it possible to return web pages which have not actually been crawled.

    The logic was sound, and the early results impressive, however…

    This was VERY open to manipulation.

    To rank a web page for a query, all an SEO had to do was point multiple links at it with their target keyword as the anchor.

    More keyword rich anchor text links than your competitor = you won that particular rankings battle.

    This also lead to some amusing examples of ‘Google Bombing’ — where SEOs would show how easy it was to game Google by pointing anchor text links at non relevant pages and ranking them for the lulz.

    This 'Google Bomb' successfully ranked George Bush for "miserable failure" for over 3 years

    One ‘Google Bomb’ successfully ranked George Bush for the term “miserable failure” for over 3 years

    Clearly things had to change.

    Google Fights Back Against Manipulative Anchor Text

    Since us SEOs can’t have nice things without spamming them to death, Google gradually began to reduce the weighting of anchor text in their algorithm.

    But it wasn’t until 2012 that the Google Empire really struck back.

    empire-strikes-back

    In April of that year the first iteration of Google Penguin rolled out and hammered low quality and manipulative link building.

    penguin

    Anchor text was one of Penguin’s primary targets, and many websites that had been overly aggressive with their exact match anchor text links saw their rankings tank overnight.

    It may have only impacted 3% of search queries, but as usual, Google’s ‘Project Fear’ had done the trick.

    This email, received by Ryan Stewart of Webris from a client he built some solid white hat links for, sums up today’s twitchiness about anchor text.

    We are looking to dilute our anchor text profile. We hired you to build white hat links – anchor texts should ONLY contain our brand name! Anything else is manipulation and we will get penalized by Google.

    So is Ryan’s client correct? Should we really be avoiding using ANY keyword phrases in our anchor text to avoid penalty? And for that matter, does anchor text continue to influence rankings?

    To find out, we conducted 2 studies:

    The first study was small and looked at anchor text influence on rankings in competitive niches. The second study cast the net rather wider and looked at anchor text influence across 16,000 keywords.

    #

    I manually analysed the rankings of 51 keywords in big, high paying niches – think pay day loans, law, finance, insurance, health etc – with high keyword difficulty scores.

    Here is what I discovered.

    Influence Of Exact Match Anchors

    First I took a look at the percentage of exact match anchors from individual referring domains — calculated against the total number of individual referring domains to the URL.

    I included the median figure as there were certain niches (for example payday loans) that were very anchor text heavy, so I was conscious they may distort the percentages.

    exact-match

    Now for the actual volumes (number) of exact match anchor links.

    volume

     

    INFERENCES

    The data above would suggest that exact match anchors still have some influence on top placements in competitive niches.

    Average is high, but as mentioned the data is skewed somewhat by certain niches, so I would err towards the median results and suggest a percentage of 1 to 2% (and a low number, i.e. 2 or 3) would be safe and potentially help with rankings at page level.

    Influence Of Phrase Match Anchors

    Here are the average and median percentages of phrase match anchors by position.

    This is where the search phrase appeared in its entirety within the anchor text(including exact match) and again is calculated as a percentage of the total number of referring domains to the URL.

    phrase-match

    And the volumes…

    phrase-volume

    INFERENCES

    The data above would suggest that anchor text containing a target phrase continues to influence rankings.

    Average and median figures are closer than with exact match, and it would appear that around 30% of anchors containing a target phrase would be a desirable and safe level.

    It would also appear that a fairly even split between DoFollow and NoFollow correlates with rankings.

    Number Of Domains With At Least 1 Keyword Anchor Link (Phrase Or Exact)

    The following data shows the number of ranking domains containing at least one anchor text link (split by phrase and exact).

    one-anchor-link

    If we look at SERP position 1, it becomes clear that there is a correlation with a page’s link profile containing at least one keyword rich anchor and its ability to rank highly for that keyword.

    no-keyword

    #

    After completing my initial study, we wanted to tap into Ahrefs HUGE data and scale things up to see if the takeaways would remain the same.

    This time we looked at the top 20 results across 16,000 keywords — meaning we analysed a whopping 320,000 pages in total!

    For each page (URL + Keyword) we collected the following metrics:

    • Ahrefs rank of the target URL
    • Total # of referring domains of the target URL
    • # of partial keyword match referring domains
    • # of exact keyword match referring domains
    • # of DoFollow partial keyword match referring domains
    • # of NoFollow partial keyword match referring domains
    • # of DoFollow exact keyword match referring domains
    • # of NoFollow exact keyword match referring domains
    • Avg. DR of the partial keyword match referring domains
    • Avg. DR of the exact keyword match referring domains

    Experiment #1

    First off, we wanted to find out if there was a correlation between the percentage of exact/partial match anchor text links in a page’s backlink profile and its ability to rank higher in search.

    Here’s what we got:

    anchor-text1

    INFERENCES

    It certainly appears that higher ranking pages tend to have a larger percentage of exact match anchor text links in their backlink profile.

    Although, perhaps surprisingly, the correlation of partial match anchor text links wasn’t too prominent.

    But that was a percentage of links from the total backlink profile of a page.

    What if we look at a raw number of referring domains with both exact and partial match anchor text links?

    anchor-text2

    The takeaway here is clear – anchor text links from a variety of different domains correlate with higher rankings in Google.

    Experiment #2

    One thing that Experiment #1 revealed is that pages with more backlinks from different domains tend to outperform pages with less backlinks.  This is what we would expect to see.

    So in Experiment #2 we decided to level the effect of a strong backlink profile and only compare the data on pages with similar numbers and quality of incoming links.

    We limited our keywords to those where the Top 5 Google results’ URL Rank had standard deviation of less than 30% of their average value. This left us with about 2,000 keywords (out of the original 16,000).

    Here’s how this changed the numbers:

    anchor-text3

    INFERENCES

    In cases where pages had equally strong backlink profiles it was the % of partial match anchor text links that mattered the most, while the % of exact match links stayed more or less consistent across the Top 5 results.

    #

    Both case studies clearly demonstrated that keyword usage in anchor text continues to influence search rankings.

    However…

    The Caveats

    Firstly, there is the standard correlation does not imply causation.

    Google has over 200 ranking factors, and while the data above would suggest that keyword usage in anchor text continues to be one of them (and you probably should build some keyword rich links) there are numerous other factors which may have influenced the rankings of these pages.

    Secondly, we are unable to tell if any of the links I/we discovered have subsequently been disavowed, which could be the case in some of the particularly keyword rich profiles.

    So as always, take everything with a grain of salt and conduct your own tests.

    That being said…

    I Believe Anchor Text Continues To Influence Rankings

    Caveats in place, I’m going to stick my neck out and say that keyword prevalence in anchor text is almost certainly still a ranking factor, and perhaps a bigger one than Google would like us to believe. It’s what the data tells us.

    You can quote me on that.

    So, if you’re building links (or someone is building links for you) then you probably don’t need to be too sensitive about anchor text, and a few keyword rich links as part of a diverse profile should help you rank.

    Just don’t go crazy though and from the data above, I would suggest exact match at around 2% and phrase match at around 30% will keep you on the right side of Google.

    Of course, as always, the strongest and safest links are those which are editorially earned and you should always avoid building low quality links, regardless of anchor text.

    Over To You…

    What’s your thoughts on keyword rich anchors?

    Do you continue to build them, or do you avoid them like the plague? Any observations from your own site? Have you been penalised in the past? Any other questions?

    As always it would be great to hear from you, so please drop a comment below!

    David McSweeney
    David is the blog editor here at Ahrefs, the owner of Top5SEO and a white hat SEO evangelist. SEO case studies make him a lot happier than they should, and he has a tendency to overuse ellipses...

    Get notified of new articles

    46,961 marketers are already subscribed to Ahrefs blog. Leave your email to get our weekly newsletter.

    • Hi David,
      Many of my confusions on this subject just washed away by going thru this post. What an information packed post. I just book marked it for my further read and study. Hey, I am so glad to say that I just curated this post in some of the top curating platforms like Inbound etc. The connected links give further information to study on this subject. In fact I was so ignorant about this subject
      Thanks for clearing my doubts.
      Keep sharing
      Have a wonderful rest of the week
      Best
      ~ Philip
      @PVAriel

    • I agree…

    • I’m doing linkbuilding for several (big and small) clients and your percentage is quite near the percentage we’re holding on to (3% and 25%). Nice to finally have an article to back me on up on this. Thank you for this!

      • eugene

        hi just interested to know what type of backlinks you still build on? thanks.

        • That question is quite broad. Do you mean types as in in-content links, local links, directory links, social links etc? Or do you mean follow, do-follow? Or perhaps exact match, branded, generic and naked urls? The answer is: all of them, if needed. 

          I first look at the current backlink profile of the page/website of the client and base the linkbuilding on this.

    • My friend César Aparicio will love this post 😀

    • thanks my friend .I am following you very beginner leval. After read this article manu confusioins clear about anchors. you are my man brother. hope you’ll continually some new advamce techniques. 

    • How does one go about getting access to that much data?

    • Jason Barnard

      The word I “hear” here (!?) is “Balance”
      Often the right word in SEO

      • absolutely! The problem the SEO community often has is that if something works, everyone takes it to the extreme… and then… it doesn’t work anymore.

    • Kuba Kzz

      I always try to include the keyword in a more diluted anchor phrase — like 2/3 words into a 5 word anchor. 

      A nice trick is to come up with a short tagline (containing the most important keyword) that comes with the name of your brand eg. “Brand — adjective keyword”. 

      Then you get more links with the keyword, and at the same time they (in my opinion) safer as they contain the brand name and they are dilluted.

    • Wade Cockfield

      Hi David nice article, thanks for all the research. I’m a bit confused by the closing recommendation of 2% exact and 30% phrase. I read the graphical data for the No1 rank as: exact at between 11–13% and phrase between 22–24%. ?

    • Thank you for sharing! Data enables professionals to improve their individual standards in an industry where standards are all over the board! Raw data helps us innovate and overcome challenges! I look forward to more posts like this.

    • Michael Mahoney

      Good stuff, thank you! Is it safe to assume that there’s a different formula for interlinking anchor text within the website itself?

      • We only looked at external anchors, but I think you should be pretty safe with what you do on your own site. Probably try and mix it up though.

    • pokemon

      Ohh, look amazing post. Thank you David.

    • Nigel Griffiths

      Very interesting post.…thanks David. This is why we love Ahrefs — the more information the better!!

    • Thanks for article! Will use it for my sites.

    • What an awesome post, I just read it from start to end. Learned something new after a long time. Thanks 🙂

    • NCode Technologies

      Aweson article David. It is so detailed and well formatted that i enjoyed reading it as well as get some new information too. It will help us to diversify our anchor text portfolio too. Thanks.

    • Great detail. Anchor text has been challenging for me and I’ve learned some hard lessons. I wish I would have had this post a year ago, what a difference I could have made. Thanks for sharing!

    • I really enjoyed this article Tim — good work 🙂

    • Nice article David. And as Tim says Ahref does post fewer articles. So much so that I find myself checking my mail box every once. But my question is how donI control how ppl link back to my site, especially since I would like to earn my links on the back of great content.

      • If you’re getting editorial links to your content, then I wouldn’t get too hung up on anchor text. There are certain times when you can control the anchor though and I guess that’s when you might want to consider using your target phrase.

        • Oh ok. I thought I was missing something here. Thanks for clearing up my doubt.
          Regards.
          PS: I hang on to every word from u guys!!!

    • David,
      That was heavy reading and I’m no slouch. Thank you for the in-depth research and the level of detail. The takeaway of 2% exact and 30% phrase is very helpful. Sometimes I need someone to spell it out.
      Thanks

    • The permissible anchor text density also varies throughout every industry as well. In some niches like ‘Web Design’ I see the top 5 sites all ranking with 20% + exact mach anchor text! In other niches it’s a lot cleaner and every site on the 1st page has 0% exact match for their main keyword like ‘Psychologist city’. Every industry / niche has different levels of acceptable exact match ratios.

      • One of the issues there is that “web design” is an actual phrase that people would use, whereas “Psychologist Glasgow” for example is not something people would actually say — other than when they are searching. So if that phrase does appear in text, it’s going to look forced. My guess would be Google is smart enough to know when something is not a real phrase.

    • oh really a great brief info on anchor texts,so if there is exact match anchor text then it is not good ? which is what you want to say right

      • depends where they come from and how many there are. The data showed that some exact match anchors correlate with rankings.

    • thanks for article David McSweeney

    • Esther Max

      At the beginning of this article you set the terminology stating “partial match” — is this the “phrase match” you talk about?

      You got me confused
      What is the point of stating a terminology if you don’t follow it ?

      • Hey Esther,

        The answer is yes… and good point! I have updated the definitions to include both phrases.

    • John Stover

      I try to only use exact/partial (mostly exact) anchors with links, which push a ton of authority. Leave the low end links (foundation type links..etc) for the brand stuff. Works miracles. There are a bunch of other variables, including context and how you funnel your links, but from a best practices standpoint, this is a good start.

      Great article David!

      • sounds like you’re doing it right John and glad you enjoyed the article 🙂

        • Andy

          Hi David,

          You mentioned that for study 2, you collected data on the average DR of exact and partial keyword referring domains, but you didn’t elaborate on how much impact this had on the rankings (i.e. quality of anchor text links, rather than quantity). John Stover’s comment implies that DR (and therefore quality) does matter, but I’d be very keen to see some data behind the theory.

          Awesome article David. Thanks.

    • True. Great Article sir. Thanks for sharing.

    • Great article, David! I really enjoyed it, and I love the data-backed findings. I guess the challenge here is that partial or exact-match anchor text is an easy red flag to alert editors, who are often now well-trained to spot subtle SEO maneuvers like this. 

      When an editor suspects you of manipulating your anchor text for SEO, I’ve found that they’re far more likely to remove your link, nofollow it, or, in some cases, discontinue working with you altogether. 

      So there’s a careful and difficult balance. Every time you use partial or exact-match anchor text you’re taking on some significant risk. 

      Of course, this risk could be negated if you just focus on getting editorial links without worrying about anchor text, and then use internal links (ie, links on your own website’s pages) to pass anchor text relevance. Do you have any data on whether internal links’ anchor text plays a role? And if so, how much compared to external link anchor text?

      • Thanks Jayson.

        I definitely agree that editorial links are the way to go, but when you do have the opportunity to ‘manipulate’ the anchor, then it’s certainly worth trying to squeeze a few partial matches in at least. And to be honest, I prefer partial match to exact match anyway as I think it’s a lot safer overall.

        We don’t have data on internal links yet, but I certainly always try and link contextually as much as possible without going overboard. It’s a question I’ve seen come up a lot lately, so perhaps it’s something we’ll try and look at in the future. I guess the problem would be isolating the data (one for the data scientist!).

        • I’d love to see that analysis (internal anchor text)!

    • Thanks for detailed information about Anchor text.

    • Joelee

      Thank you for clarifying the muddle at times.

      I have one question which I’ve had conflicting replies to when placing the question direct.

      The anchor text ‘dog busicuit’ as the example. I have my site which has a page called ‘dog busicuit’ . If I have Linking into that page is useful, but are there any pros and cons to linking to the page if say the H1 is ‘dog buscuits’ or to multiple text anchors on that page?

      Is there any value to my page linking to dog busccuits? Or is the value solely to the receiving anchor.?

      Thanks all

      • Hi Joelee, sorry, I’m not quite sure what you mean here 🙁 Do you mean mis-spellings?

    • Alex@TheBeach

      David, great article thank you. For clarity’s sake I’m confused about how you used partial/phrase match anchors in your test. If your target phrase is “widget shoes sale” and the anchor is just “widget shoes” or “shoes”, were you counting those as partial/phrase since they don’t have the complete target phrase in them?

      Thank you.

      • Hi Alex, in this case it was phrase match, so for example “find a great widget shoes sale here”.

        • Alex@TheBeach

          Great to know, thank you David 🙂

    • Fantastic post, I really needed this guide to clear my misconceptions regarding anchor text usage. You cleared my all misconceptions. Thanks bro 🙂

    • Nick Carter

      Wow, thanks for data!

    • Thanks for the information and the API link

    • vov David McSweeney! thank for post.

    • Hi David

      Fantastic article, thanks for all the research. I use them often but going to adjust my strategy after reading this.

    • Such a great article. this article clear my doubt on anchor text.

    • Hello, Kindly give some update after latest changes and penalties from GOOGLE in November & December 2016

    • Love The Anchor Text.
      Thanks for the Blog David.

    • Paul Slater

      Great article. One question I was hoping to get answered though but didn’t see is what about anchor text for email addresses on a website. Does Google punish you for having anchor text for an email being the email address itself? E.g. myemail@email.com. Or should the anchor text be Contact Me which when clicked on opens the user’s default email client with the email address in To field ?

    • Got here from your ‘Case Study 1: How We Increased Search Traffic 51% Using ‘The Merger Method’, David… LOVE! Love the case study, love what you did with this guide — happy to see others love it as well.

      Question: what plugin do you use for your ‘Quick links’? Thanks!

    • This is why I’m an Aref fan. Always enjoy the detailed analysis. Nice work, McSweeney.

    • A very interesting guide that is a good complement for the last article about this item.

    • Interpolyst

      Where is the naked URL and branded in that story? Or I’m mixing pears and apples.

    • Thanks for information about Anchor text.