[MarkAse]

I personally haven't seen evidence of the anchor text problem.

My brick/mortar targets over 250 keywords and has a few thousand blogroll links linked to my company name: killed.

A couple of local rental sites have 100% of the anchor text pointed to city + service.  Literally 100%.  Unchanged rankings there

[Joe Hughes]

mikemunter, on 26 February 2012 - 13:13, said:

Joe - I have a hard time believing it is an "on-page" penalty because EVERY keyword I was building links for prior to Feb 15 was ranking very well on the first page of Google.  Just by the odds, at least a COUPLE of my pages should have not been hit, if by luck alone.  Could on page be improved?  Sure, but I can't just sit back and say it must be on-page alone.  The #1 site outranking me now for my main keyword is garbage - the guy's site is stuffed with keywords at the top and bottom of the page and every single page on his site follows a template with only minor changes.

My anchor text was diversified and I had many different pages ranking on 2 different sites.  They all have dropped overnight.

An area where I personally may have failed is only having about 5-10% of anchors being things like "site name", "URL", or "click here" type anchors.  Obviously, in the "natural" linking world, I'd had a much higher % of links like these.

Can anyone else comment on this anchor text link diversity?  Can anyone comment on sites they have that have not lost rankings vs. those that did and any differences you notice for your linking strategy?

When we talk about Panda and on page we're not talking about page titles, keyword density etc. The on page is what Panda went after originally. They tried to determine if your site was "trustworthy" via the algo.  So they looked at duplicate content, page layout, over lapping content (a lot of people got hit because of this) and a whole bunch of other signals.

My view is this was pretty succesful on the original update but the Feb 2012 updates have hit a lot of quality sites particularly in the ecommerce sector.

[PBR]

I throw tier 2 lnks at BMR posts on high PR root domains and so I keep track of them. Since December I have noticed them (urls) being dropped from the index at an increasing rate and this got me working on replacing BMR with other things. I believe that this negative velocity in powerful links is one factor at work here.

Another factor is an algo change this month which I believe has reduced the potency of BMR links to boost SERP ranking. My experience is that the more you used BMR in your mix, the more impact you will see (although the picture is fuzzy because nobody outside of G knows how these things are worked out).

A final issue is that there seems to be some internal problems at BMR. Posts are being delayed and the new quality criteria are being implemented in a patchy fashion. I have switched to 100% American writers and I am still getting posts rejected. When I look at the content on the network and I see some of the crap that they are currently posting, I cannot understand why I am getting perfectly good posts rejected. This means that the economics of using BMR is something that I am going to review at the end of the month with a view to potentially dropping my subscription.

On the positive side, BMR links still work well in the early stages of ranking new sites. The questions are whether it is worth the money and whether it will be around in 12 months.

[kschmandt]

Joe - I don't see how this could be an on page panda issue. All of my posts are over 1k words, all have good content, well written, outbound authority links, not just all aff links, Images, pros and cons very clear easy to follow layout, and a very low (for a review site) bounce rate of only like 50%. I just don't see how it could be an on page issue.

I guess the only thing I didnt mention is that I have only really been working on this site for a month or so. I bought it maybe 6 months ago it had just expired and I really liked the name. So the domain is 3 years old and I put a few posts on it several months back but I have only been actively working on it in the past month. Only 2 weeks ago when G updated PR I got bumped up. I have to have gotten some kind of linking penalty its the only thing I can figure.

[live&learn]

Hey,

I thought I'd chip in to the discussion as what I've experienced is similar to what others have experienced - high first-page rankings obliterated (seemingly) overnight. I'm talking about plummeting from #4 to #650.

I do use BMR links and a lot of my "high-powered" (i.e. high PR) contextual backlinks to this page came from BMR. I checked about 10 of my BMR posts (to see if they were still indexed in Google) and all of them were. However, that's probably not a large enough sample to be sure - I had about 160 BMR posts linking to the page in question, and don't know how many (if any) of those posts were deindexed.

But i have to say, I've always had concerns about using BMR. I remember writing to "John" (it must have been the same John you other guys mentioned) and expressing my concern that Google could surely detect (and deindex) their sites very easily.

As someone else said (in this thread or another like it), it's just a matter of signing up to the network:

(i) Google get one of their employees to sign up for BMR (they obviously know about BMR as BMR's site is in their index and they've got whole teams devoted to spam, search quality, etc.)
(ii) That person creates a ton of posts
(iii) That person then finds out which domains the posts were published on (and notices other suspicious looking posts there too)
(iv) (S)he reports back to Google and those domains are deindexed

Surely it's that simple. Detecting the network is hardly rocket science. I mean Google are employing human beings to review sites, now, so I suppose there's a chance they also employ people to detect link networks. Any thoughts?

[neil_patmore]

Another network seems to be biting the dust.

[mofoe]

live&learn, on 04 March 2012 - 12:50, said:

Hey,

I thought I'd chip in to the discussion as what I've experienced is similar to what others have experienced - high first-page rankings obliterated (seemingly) overnight. I'm talking about plummeting from #4 to #650.

I do use BMR links and a lot of my "high-powered" (i.e. high PR) contextual backlinks to this page came from BMR. I checked about 10 of my BMR posts (to see if they were still indexed in Google) and all of them were. However, that's probably not a large enough sample to be sure - I had about 160 BMR posts linking to the page in question, and don't know how many (if any) of those posts were deindexed.

But i have to say, I've always had concerns about using BMR. I remember writing to "John" (it must have been the same John you other guys mentioned) and expressing my concern that Google could surely detect (and deindex) their sites very easily.

As someone else said (in this thread or another like it), it's just a matter of signing up to the network:

(i) Google get one of their employees to sign up for BMR (they obviously know about BMR as BMR's site is in their index and they've got whole teams devoted to spam, search quality, etc.)
(ii) That person creates a ton of posts
(iii) That person then finds out which domains the posts were published on (and notices other suspicious looking posts there too)
(iv) (S)he reports back to Google and those domains are deindexed

Surely it's that simple. Detecting the network is hardly rocket science. I mean Google are employing human beings to review sites, now, so I suppose there's a chance they also employ people to detect link networks. Any thoughts?

OK did you overdo the anchor and did you overdo the links to the index page only?

[live&learn]

Hey mofoe - if you mean did I create too many links with the keyword as anchor text, then I don't think so. I've heard people say that you don't want to overdo it, however I also know of cases where people have used their keyword as anchor text almost 100% of the time and they rank just fine.

Terry (Kyle) himself recommends 70%-80%. I'd say I was hitting the 80% mark (keyword as anchor text 8 times out of 10).

As for overdoing links to the index page, can you explain what you mean? Do you mean "was I linking only to the page I wanted to rank?" or something else..?

[live&learn]

Scrap (some of) that - just read Terry's post. Looks like the latest best practice recommendation is to use only 50% keyword-as-anchor text, so perhaps that's one of the reasons why I was hit (in addition to using BMR directly on my money page)

[mofoe]

Anyone please before you go and post in this thread, infact go and do it now. Please use a backlink tool to check your % anchor text and THEN report your results. I suspect an awful lot of you are moaning about BMR but are guilty of getting an OOP.

[trakker]

mofoe, on 05 March 2012 - 02:16, said:

Anyone please before you go and post in this thread, infact go and do it now. Please use a backlink tool to check your % anchor text and THEN report your results. I suspect an awful lot of you are moaning about BMR but are guilty of getting an OOP.

Good point mofoe, could be a factor.

@liveandlearn - you don't think 80% is overdoing it?  Really???  You think that looks anything close to natural???     Yes, some sites get away with near 100% keyword use, but that doesn't mean its smart, and that you'll be lucky like that too.  Even if it does work for them or you today, probably won't tomorrow.  

I don't know how much such over optimizing anchor text has played in the recent drops people are seeing, but they  Google has definately changed the algo to not reward such silly unnatural linking.

I've always used less than 40% keyword in anchor text.  But have been trading some links with others on my network and recently noticed those that requested single anchor text have not done well, while those that asked for variety of anchor text rock.

Edited by trakker, 05 March 2012 - 05:22.

[mofoe]

Trakker I have seen it and pretty sure that what Matt Cutt's means by algorithmic penalties coming back from the dead after the recrawl if you have "sorted those links out"

Also we have seen countless webmaster penalties notifications on here about "unnatural linking" . Hmm wonder what could trigger that....Too much anchor text!

[abby013]

Hi guys,

Just want to let you know that the latest algo was:

deindexing hpbl and does not vary your anchor text....

for those who have been hit, used 5-8 varied anchor text and i can guarantee that after 2 weeks you will see improvement on your ranks...

[abby013]

why hpbl are deindexed?
easy... domains with page ranks hosted on class C ips pointing to a same site..

why bmr is deindexed?
easy.. blog post hosted on class C ips pointing to same site...

for those who have used bmr, aln or any private blog post network.....
if you observed sudden drop of keywords ranks, there are 2 possible reasons...
your website is reviewed (if your website is an affiliate, you will fail on G review) or you use 1-3 anchor txt/url....

Edited by abby013, 05 March 2012 - 07:23.

[gregmac037]

just follow what abby says shes right.....

Edited by gregmac037, 05 March 2012 - 07:33.

[GeorgR.]

live&learn, on 05 March 2012 - 00:55, said:

Scrap (some of) that - just read Terry's post. Looks like the latest best practice recommendation is to use only 50% keyword-as-anchor text, so perhaps that's one of the reasons why I was hit (in addition to using BMR directly on my money page)

Yes of course - did it STILL not ring through to some people what "link diversity" actually is? This ALSO means varying anchor texts.
Look, look on the "real" web and see how many "real" links are keyword/anchor text links. Fact is: The majority of natural links are NOT anchor text links.
And no matter what link building you do, you have to make it look as natural as possible.

[live&learn]

@trakker - well, I certainly now think that 80% is overdoing it. However, it's worth pointing out that Terry himself recommended up to 80% anchor text in SEO masters academy (and this is someone who's successfully ranked hundreds of pages). Plenty of other SEOs have recommended a similar figure too (Matt Carter and so on).



I think hindsight is a wonderful thing. Certainly 80% doesn't look "natural" but if it's what successful SEOs have been using to achieve top rankings (and what they were recommending that others use as well) I don't think it was that crazy to give it a go. That said, things seem to be different now. In Terry's big post, he revised the figure to 50% (if I remember rightly) and that's probably what I'll aim for in future.

[live&learn]

@GeorgR - point taken. However, what I said to trakker applies here too: Yes, "natural" linking on the web does not make such repeated use of the same anchor text. But high percentage keyword-as-anchor text has worked in SEO for some time (again, I refer to Terry Kyle's own previous practice and recommendations of 70-80% keyword-as-anchor text). And as long as it worked (it worked for me a few times as well) there was not much reason to change it.

That said, I'm now certainly going to change it - 50% at the most, probably. I do agree that we want to make our linking look "natural" and will be making that more of a priority from now on.

[John Moore]

The BMR saga just re-enforces the importance of using multiple linking strategies/networks to build your link profile, even though BMR deny being hit hard, the fact is I know a lot of loyal users all complaining of ranking drops and their posts not being indexed any more.....since they're probably the biggest SBLOG network out there they'll always have a big red target on their heads from Google....