Backing Up The Brand – Are Over Optimised Link Profiles A Barrier To Top 5 Rankings?

Well it seems like my last article caused a bit of a scare and sent shivers down the spines of link buyers everywhere πŸ™‚

No, but seriously it had over 10,000 hits last week and was reported in all the right places, however there was the odd murmur labelling the article unreliable and structured poorly. They may have had a point but I think I made it clear at the beginning of the post that it wasn’t scientific in anyway.

There was one particular point that seemed to hit a nerve and that was the point about branding or more specifically how a more natural, brand targeted anchor text profile could be getting more ranking weight than usual. This point of view was based on experience rather than in depth research, however I thought I’d look into it a little closer.

Please note: This is personal research performed over a limited number of search results and keywords. The research shows correlation between a sample of link profiles and their rankings for a given keyword, it does not establish a guaranteed ranking factor

Ok now I’ve got that out of the way let’s jump right into what I did.

The Method

I chose 10 keywords that I monitor closely for personal and business reasons;

1- perfume
2- search engine optimisation
3- internet marketing
4- loans
5- car insurance
6- injury lawyers
7- search engine marketing
8- flights
9- car finance
10- property for sale

The top ten results for each of these keywords were then examined and their link profiles broken down. Only links from unique domains were looked at and the relevant anchor text, the two types of anchor text in question were the exact match keyword and any brand related terms.

Exact Match – Only keywords that match the search query exactly

Brand terms – All terms relating to brand e.g., brand,,

Once the links were collected the percentage was calculated compared to the total number of unique links.

The aim was to see how brand heavy link profiles compare to heavily optimised anchor text.

The Results

The results I found confirmed some of the ideas I’ve had about anchor text and how a more natural profile is now receiving higher rankings. Either that or some of the obviously manipulated profiles are getting filtered out.

Anyway take a look at the results, 10 different keywords, 100 different websites and here is how the link profiles turned out;

brand anchor chart

As you can see from the trend lines, the sites with more brand or URL links within their profile, generally rank higher, at least across this set of keywords.

In some industries the websites at the top had far less linking domains, far less exact anchor text and yet still out ranked the some really strong websites, for instance take a look at the perfume keyword;

perfume anchor chart

The trend lines becomes even more aggressive, it’s almost a barrier to top 5 positions to have over optimised link profiles.


Ultimately this may have nothing to do with emphasis on brands, instead it might be a simple filter on over optimised websites, all I can do is fall back on what I said in my previous post.

Around April/May there were some huge ranking swings across the board, there were various updates to top ten positions, from my experience and the little bit of research I have done it is due to the above. As SEO’s and webmasters we need to start giving more attention to a natural link profile, yes we still need a portion of optimised links, however in my opinion that portion is nowhere near as important as it once was.

For a long time now optimised Anchor text has been the SEO’s ultimate weapon, in some niches it was enough to dominate the top 5 positions, for me the latest swing in rankings makes it clear that Google is going to be more on the ball when it comes to discovering over optimised profiles and more willing to filter out websites that on the face of it look strong but underneath the hood highlights 99% manipulation. Try and discover ways for your site to earn natural links and then compliment that with some old school SEO.

To those who claimed my last article was β€œthe reason the SEO industry has a bad name”, how was this one? Your feedback is always appreciated πŸ˜‰

As always thoughts are welcome….

You should also check out this awesome break down by Nichola about over engineering the link graph – Really great read!

p.s. Thanks for all the emails about the malware attack, it is now clean and the hosting issue has been solved, it was cleaned up within an hour of infection.


Tim Grice is the CEO of Branded3, a search marketing agency in the UK. Tim has over a decade of experience in search and regularly speaks at key events and conferences.

Latest posts by Tim (see all)

51 Replies to “Backing Up The Brand – Are Over Optimised Link Profiles A Barrier To Top 5 Rankings?”

  1. Tim I hope your next post is going to give us “solid” examples of how “brand” links should look.

    I’m not sure if I have a specific problem because my 2 main sites are general article directories. So am I to try and encourage links that are anchored to the name of the directories or the url of the directories?

    I’m pretty sure I understand what you are saying but I’m just a bit unclear on how I should proceed.

  2. Hi Pchef,

    This change up in the SERPS, in my opinion, is down to 1 of 2 factors.

    A) Google is placing weight on links anchored with either the URL/Brand Name/Other variation of URL e.g. seowizz, seo wizz, http://www.seowizz,net, etc….

    B) Google is filtering sites that have an over optimised anchor text profile, for example if 60% of your links are anchor with the exact keyword then this looks very unnatural, in order to combat this you should mix up your anchor text and add natural link text like the url/name of site.

    When I talk about branding all I am talking about, in this scenario, is a natural link that has the anchor text of your domain in one of its variations. With deep pages in an article directory I would try and use 20% target keyword and the rest attempt to get the top level domain name plus the category.

    e.g. my article site + home improvements

    When building your link profile I would always say to myself do I have enough natural looking anchor text to show this is a well linked to site and not just a well optimised profile.

    Sorry for any confusion, Google tends to throw these curve balls every now and again.

  3. Hi Mark,

    No I purposefully filtered this out. Just the variations I mentioned url or exact url keyword.

    I am not sure if it is a brand related issue or a filter on the anchor text, however it definitely shows the need to neaten up the link profile.

  4. A right mix of anchor optimized links and brand links can give your better ranking. Change your anchor text for links once in a while. If we try to dominate the search with anchor optimized links, search engines will surely filter out the results. Don’t over optimize let the webpage/website gain some natural links. Regular update on the website is going to keep the website fresh and search engine friendly.

  5. Hi Emory,

    We all noticed a rankings switch around this time and after quite a bit of research were confident, link buyers, footer spam and anchor manipulation have all been targeted in this update.

  6. Thanks for the input.

    I think people should still try to target their anchor text but keep it reasonable and make sure they have plenty of ways of developing natural links.

  7. I think the heat you took on SE round table was unfair. I have noticed this type of trend with my agencies’ clients as well. It looks like natural looking links are becoming important again. The days of only building non-branded anchor text may be far and few between.

    Do you foresee the rise of directories and other easy ways to build branded, natural links? If not directories, what sources to you see becoming more popular?

  8. Hi George,

    I expected a little backlash once it went viral, mainly because I never intended it to get so much attentions. To be honest we have seen this trend across all our clients and have done some consulting on this exact issue with 2 of the UK’s biggest online brands. Even though they are online brands they have saturated their anchor profiles leaving them open to the improved filter.

    We personally think that Google are getting so good at devaluing links that look like they are paid i.e. footer, side bar , site wide etc…. that things like article marketing, blog writing and directory submissions are having a much bigger impact on rankings, still it will only change again in 3 months πŸ™‚

    Regarding all those who said the article didn’t show enough evidence, well, their right. I don’t let everything out of the bag people can either trust or ignore, in the meantime we’ll keep ranking! πŸ™‚

  9. It is quite interesting to talk about the good old times, when I had to write a personal email to some webmaster, telling them how great theis website is and that a link would be very helpful. I do not think that anybody answers to such emails today!
    But otherwise you are very right and over optimization of a keyword is bad, but I am woundering how could one check this. I see that nowadays nobody leaves a link in a forum without an anchor. I even think of turning the mod_rewrite off. We have enought url shorteners nowadays.

  10. Tim,

    I think you are spot on with this.

    I have noticed a considerable drop in my rankings I used to dominate in the past 2 mo. Investigation of my link profile and Google led me here and your previous article.

    I discovered a spike in links from a couple sites that have created side bar links to me in splogs. I believe they have competing web sites and are actually creating over optimized links to me as a black hat tactic. It seems to be working. I have dropped below 5th position for many terms I used to rank #1 for.

    Question for you is: does Google only devalue the links or do they devalue the site so that it drops in the rankings as I have? Also, what should folks like me do? Contact web masters and ask them to remove me from their side-bar? Hire you to look at the data?

    Your help much appreciated.

  11. Hi TC,

    From the research me and the team have been doing there are no direct penalties to websites, it is a case of the over optimised profiles carrying little weight on the terms their targeted towards. Personally I wouldn’t remove any links as this to me could give off more negative signals. We are advising clients generally to focus more and brand or natural anchor text and seems to be working well. Obviously the exact amount of work and strategy needed differs on a case by case basis, even some big brands are suffering at the hands of this tightened filter.

    Of course if you need any assistance, help or an audit just fire me a quick mail.

  12. Hi Heinrich,

    The way were finding over optimisation issues is through the open site explorer tool, pulling all the links into excel and filtering by anchor text. We have done some huge audits recently and evidence is stacking in one direction, even the biggest of brands are been hit if they are too aggressive with anchor optimisation.

  13. I’ve noticed some changes recently on some sites that I’m working on. I tried two things
    1) Link building with a mixture of primary and secondary keywords, company name and domain name. Also I made sure some were nofollow links rather than 100% dofollow.
    2) Heavy handed link building with just mainly primary target text, side bar links on a few sites that I “acquired”

    1) With last update page 1 for reasonably competitive term in London
    2) Target page has been deindexed!! Domain is fine and other pages rank no 1 but 1 page has since been removed completely where it was on page 3.

    I wonder if I repeated the steps in #2 for a competitor could I have them deindexed!? It worked for me….


  14. Interesting Dan, if you don’t mind saying what link building techniques were you using and how old are the sites? I have seen some sites penalised if they are new and you go in heavy with tons of site wide links.

    In terms of doing it to a competitor, I have never been a believer in ‘Google Bowling’. I think if a site is reasonably established and offers a genuine service/resource it is very difficult to get it penalised with link building. That been said Google are getting a lot better at detecting paid links and site wide links are paid or manipulated 90% of the time, so why wouldn’t Google punish them.

  15. Hello Tim,

    The site was created on 26/01/2010 and at first I was just doing some basics like forums, blogs, articles, press release, social etc which helped move the target page up. Recently I bought some themed links (yes I know it shouldn’t be done!) from some private sellers (wanting to experiment) so no big link networks or anything and I’m assuming this is where the problem is as since then the page was completely removed. The links are not site wide but homepage side bar links.

    I wouldn’t personally try and do that to a competitor either but I think it shows that it could be done perhaps? Although this is just one example and not enough to be conclusive at least my experiments are showing what is and what isn’t working for me at least.


  16. Hi Dan,

    Interesting, it always fascinates me when pages get de-indexed, maybe Google has started penalising sites for buying links somehow, not sure how this would work but let me know if you see any further issues or if you get to the bottom of the current issue. Have you tried removing the paid links?

    Personally I have only ever seen a handful of sites receive a Google penalty and it has usually been down to serious link spam on newer sites. Interesting.

  17. That’s my next step removing the paid links, I’ll change them to another experiment domain and see what happens. Essentially though I think my deindexed page is doomed now, I’ll let you know if I get to the bottom of it as I too think it’s interesting to find hard evidence of being penalised and what caused it.


  18. Definitely, let me know how you get on.

    At the end you could put a case study together, if the new domain gets penalised I think it will give some pretty major insights into Google penalties.

  19. Hi Tim,

    Fascinating experiments.

    Assuming your hypothesis is correct, the question it is prompting in my mind is whether it is the proportion of brand links or the absolute numbers being too low that is the key factor.

    Did you look at absolute numbers of brand links as well as proportions?

    If so, did the lower ranking sites have lower absolute numbers of brand links than the higher ranking sites or just lower proportions?

    The reason I ask is that Google’s approach, according to my understanding (“heavily underlined”), is usually to discount / devalue suspect practices rather than positively penalise and there is a big difference between those two things.

    The former approach is more likely simply to cancel the effect of hostile action i.e. neutralise it rather than penalise an innocent victim.

    If the lower ranking sites also had lower absolute numbers (and quality etc.) of brand links then that would all make sense to me. If, however, some of the lower ranking sites had higher absolute numbers of brand links (albeit lower proportion), then that would certainly create a case for further investigation (please :-).

    Hope I haven’t missed anything obvious here. It’s late over here ..

  20. Thanks for that, some really great questions.

    The research is based on proportions not absolute amounts, however our current thinking is Google is deciding weather to count a link based on the link profile of the site receiving it. So if you have an over optimised profile or disproportionate amount of anchor text, then more anchor links may only serve to worsen your rankings, where as sites with a high proportion of brand links will get more value from the small amount of anchor text it has.

    The research continues!

  21. Pardon me if being stupid but what exactly is a brand link? Is that just the url of my site or name of my site? If I want to rank a subpage, how will a brandlink help that page?

    Example, If my site is:, and i want to rank a subpage about shovels. Then the anchors should include variations of shovels. Linking or gardening to shovelpage would do much good i assume?


  22. Hi Johan,

    Yes a brand link is simply a variation of your domain name. In order to protect your website from anchor text devaluation and potential filters it is important to make most of your homepage links ‘brand’. Your deeper pages need to be keyword rich, however having said that we are seeing with more and more clients that even deeper pages will benefit from a more natural anchor text, domain + anchor text or in your case shovels from

  23. Hello Tim,

    I thought I’d report back about my deindexed page! Being quite busy I haven’t removed the paid links or done anything but in the past few days my deindexed page has re-appeared but jumping to #3 from low down on page 3! So looks like my paid links have done the trick in terms of boosting the page but I have no idea why it dropped out initially.
    Have you seen this type of thing before?



  24. Hi Dan,

    I’m dealing with it at the minute.

    I had a clean up of some (in my opinion) dodgy links pointing to my site. I have lost my search engine optimisation ranking all the way down to 90th! I do think its just the dance playing out it just pains me to lose rankings by doing the right thing. Whenever sees an influx of backlinks to a site is quite common to see the google dance play out, first page one minute second page the next 200th after that and then back to the top. It is almost like Google is penalising you until it realises what to do with all these new links.

  25. Hi, I just checked the keyphrase SEO and the top ranking site (behind Wikipedia) has less than 2% brand links to the homepage. However, the 3rd and 4th ranking sites have lots of brand links and in theory, shouldn’t ourtrank other sites below them, which backs your theory.

  26. Hi,

    Ye there are more factors involved here, however it’s not difficult to see that anchor text is not always what the rankings turn on.

  27. Thanks for this post, really could explain some of the behaviour I’ve seen from a few of my young sites – bouncing around like ping pong balls.

    Thanks for the heads up.


  28. Awesome article! I am new to SEO and this definitely helps. Does google take into consideration different Social Media accounts like facebook pages, twitter pages, linked in pages, merchant circle pages etc when ranking online branding?

  29. Hi Josh,

    I am pretty sure they do but not sure how they use the data. Certainly a site that has a FB fan page, TWitter and other social accounts would appear more legitimate than one that didn’t.

  30. Hi Tim,

    Am short of words to describe how excellent this informative article is. It gives me the causes and solutions of many problems. But it has still left me with a question.

    So as we can use as brand name, What happens when the domain is like rather than as such type of sites tend to be used for spam or to manipulate rankings.

    Would Google consider it as a spam domain or give it a equal chance to build brand as a domain without hashes when the link building is done with anchor texts like:
    keyword keyword keyword keyword

    Thanx for the article and ur much awaited reply.

  31. Hi Joshua,

    I would use an even variation, so with the .com, with hashes, without hashes, just keep it even I don’t think the hashes make a difference to the way you should link build.

  32. Thank for the reply. I have built near none links using my domain name and I suspect that to be the reason why the domain is not able to rank. Would apply this to my website and will report back about the results within a month.

  33. I am no longer certain the place you’re getting your information, however good topic. I needs to spend some time finding out much more or figuring out more. Thanks for magnificent info I was on the lookout for this info for my mission.

  34. Hi Tim,

    Recently one of my sites dropped rankings, but only for the 4 main head phrases in the industry.

    Keyword 1 – [dropped from #1 to 11]
    Keyword 2 – [dropped from #1 to 11]
    Keyword 3 – [dropped from #1 to 7]
    Keyword 4 – [dropped from #5 to 11]

    The site still gets lots of long-tail traffic (over 200 visits per day) but these phrases which are the generic, mostly converting terms have dropped off. It is an EMD .org domain for one of the main keyword phrases, so would I be wrong in assuming I could get caught in over optimising as I have the keywords in the domain, albeit with a hyphen?

    So could ever get penalised/filtered for over optimising for ‘car insurance’ for example.

  35. Hi Simon,

    When you say penalised, do you mean completely knocked out of the index? Or dropped rankings by a few pages?

    If these core terms have dropped 20 -30 places then it is no doubt some sort of devaluation of links, this could be based on the type of links they are and possibly your domain. For example, if you had a lot of site wide links pointing back at a keyword rich domains then they could easily be deemed as paid and devalued accordingly.

    Have you changed any domain registration details recently? This can effect the trust of a site and also cause it to drop rank.

    There really are lots of reasons this could have happened but 9 times out of 10 it will come down to a filtering out of some links.

  36. Hi Tim,

    I am still getting roughly the same amount of traffic from long tail searches but the head phrases have dropped down on average 8/9 places (see previous comment for details), e/g not a full on penalty. I have not changed any domain registrations. I have been targeting one, singular phrase keyword quite heavily though.

    The keyword has the follow anchor text details…

    69 IP’s


    1,388 individual Links

    Which leads me to suspect there is too many sitewide links as you say. I did just acquire a very high PR link as well which may have also caused the problems. As you can see I was #1 for 3 of the main phrases in the industry, maybe I just over cooked it.

    I can either…

    -Remove the high PR / sitewide links
    -Go for more branded links (although remember it is an EMD for one of the other phrases)

    or do both.

  37. Hi Simon,

    Unless the high PR link is an obvious paid link or very spammy I would leave it where it is, pulling links even site wide/ paid ones can sometimes make things worse as Google takes into account link growth and loss velocity. I would combat, what seems to be a filter, by building lots of variations using in content links, guest blogging, articles, press releases etc…. You need to mix up the profile and be prepared to weight. Even with a really aggressive clean up campaign filters can take 4 – 5 months to be removed.

  38. The link is pretty obviously paid, and it also has a link to the donation page. I have other sites on there not affected.

    Good idea regarding building more contextual links. If I am honest the overwhelming majority of links are focused on 2/3 phrases.

    I notice the site ranking number one now has hardly any high PR or industry relevant links, just about 500 contextual / blog links. You know the type I mean.

    Anyway thanks for the advice and ill keep you posted with the results.

  39. Yeah, contextual links can really make a difference. Get 20 – 30 posts live in one go and if you don’t see some ranking boost then the problem is definitely some sort of filter. Unfortunately it might mean a few months of varied link building including brand, this will help dilute the anchor text and allow you to start targeting your core term again.

  40. Good post! Google puts increasing emphasis on the detection of unnatural linking profiles, which is associated with the imposition of the filter. Time for a change ladies and gentlemen engaged in linking brand pages.

  41. ummm this is a lot of conjecture… which is fine because you at least make a point of letting people know. As someone who was around before the Gorg… absolutely link text changed because of the value Google was placing in it. David Harry and I have seen this and been debating a strategy change but like some here haven’t really nailed down how we change things. I’ll say this naturally occuring link text is usually not keywords a lot of times CTA like “click here” or a website name make more sense then a keyword term. In fact when Google came along I had to convince people to use keywords rather than what their natural writing used that usually was some variation on the 2 above.

    However that said… IMO, Google does not think in terms of positions so I doubt it is some filter that sets a maximum position. That’s where the post loses a bit of credibility. Is it happening? Yes I am convinved there is a filter or dampener but I would bet my last dollar it isn’t based on keeping sites out of the top 5 but could look like that…the thing is to put in terms of rankings is IMO, just not really understanding how Search engines actaually work. I think sites can still position in #1…. I’ve seen it with these link profiles however, IMO, it is wasting resources and likely a tell/red flag on bought links playing a role.

  42. I are now trying for too long to supply my
    probability of profiting in these games I am playing.
    But at any cost I does one are inclined to fail I can’t say for sure whether it is simply because the generations are growing more skilled or if I just don’t have what must be
    done anymore.

Comments are closed.