Exact match anchors, should I dilute them? Or disavow?

[zelch]

I have a site that got hit with the link spam update, I think probably because of exact match anchors (although not 100%). If there are too many links to dilute with blog type links (paid insertions/guest posts etc), what have you guys done to fix it? Disavow?

Or does it help to get a lot of lower level (cheap) links that are easy to get to dilute the anchors? If anyone has done this, what type of cheap links do you use for this?

 

[mj22]

How bad is it? How established is the link profile?

I don't use the disavow tool much unless its pretty bad, if it is ill ask the client to go in and disavow the lower end and/or sketchier looking links that are using exact match and partial match anchors that influence the main money keywords we are trying to rank for on that page. While doing this I also run either niche edits or DA30 guest posts (lower cost quality stuff so we can get more unique rd's with unique natural, generic, and branded anchors per money spend) to water down those anchor distribution percentages. Once i feel the anchor profile is diluted enough i will run some higher end stuff(DR70+) at it using branded and natural looking anchors to get some higher trust and authority toward the target url.

I hope this helps.

 

[Ryuzaki]

My advice when this comes up is to try to dilute it if it's not that bad, because the success rate of getting anchor texts changed are going to be very low. But if the number of exact match instances are low enough, I'd try to reach out to the webmasters since it won't eat up too much time, and every victory will save you some time and money during the dilution.

I've had the discussion with a couple people over whether or not the disavow file only does what Google says it does, which is to change a link to nofollow, or if it also disavows the anchor text usage, too. Which then begs the question of whether discounting an anchor text is something that's unique to the disavow tool or it's something inherent in nofollow links. I know for a fact in the past that nofollow anchor texts could screw you up, but things can change. I've heard it argued both ways, with both sides saying "but we actually tested it, many times even".

The problem these days with diluting anchor text profiles is you should probably aim to get the links indexed, if only as a signal to yourself that the links you're building are high enough quality to bother with. And since you'll need a decent volume, you'll want to keep the price down, but the problem with that is you start getting into really low authority, low trust, spammed out links like PBNs and niche edits on hacked sites with the craziest outbound link profiles. It's a predicament.

You could test disavowing for one page you know is suffering from this and see how it plays out. It could take up to 6 months or more for Google to recrawl all those backlinks though, so if you can urge them to crawl after you submit the disavow file, that'd be good. I haven't seen an indexer that works in years, so even then it might be a matter of spraying some very low-grade spam at them like dofollow blog comments or something, which isn't nice to the webmasters. It's a predicament!

 

[mj22]

I've had the discussion with a couple people over whether or not the disavow file only does what Google says it does, which is to change a link to nofollow, or if it also disavows the anchor text usage, too. Which then begs the question of whether discounting an anchor text is something that's unique to the disavow tool or it's something inherent in nofollow links. I know for a fact in the past that nofollow anchor texts could screw you up, but things can change. I've heard it argued both ways, with both sides saying "but we actually tested it, many times even".

I am glad that you brought this up, I dont use the disavow tool very often. Im usually in there disavowing shady pbn looking and bad neighborhood backlinks that come in. Other than that for penguin issues where pages are really anchor heavy i will do as i posted above. I am not sure if the disavow tool just nofollows the link or if it in fact does dismiss the anchor as well. I really dont know, i havent had a chance to do any isolated testing, like i said i dont have much first hand experience with disavowing stuff outside of what i posted.

I have recovered 20+ targets over the last couple of years with a 90% success rate doing what i stated. If the disavow tool is a waste of time i would really like to know, cuz it seems like every other month some of the agencies we supply and I consult for are sending me projects with penguin issues that I am having to clean up and its a pain in the ass. Im dealing with this on a regular basis. I guess my thoughts on it were trying to come at the problem from both ends. ¯\_(ツ)_/¯

 

[Ryuzaki]

If the disavow tool is a waste of time i would really like to know

I’ve recovered sites from Penguin penalties in the past that were spammed by automated networks, solely by using the disavow tool. It definitely works to negate links. There weren’t any anchor text issues in these cases, so I’m not really aware.

We’ve known not to hammer exact matches for such a long time I only ever see it on churn and burn projects or where people either don’t realize that’s the result because they see other sites ranking at the top with 90% exact match for a half a year until the next data drop whacks them. In some niches, the top 2 or 3 aren’t even available to anyone not willing to burn their site that way. So what happens is everyone doing research always sees anchor text over-optimization working at any given time in these niches but they don’t wait long enough to see the complete reshuffling of the SERPs.

 

[zelch]

While doing this I also run either niche edits or DA30 guest posts (lower cost quality stuff so we can get more unique rd's with unique natural, generic, and branded anchors per money spend) to water down those anchor distribution percentages.

do you have a recommended provider for the niche edits or DA30 guest posts?

I have recovered 20+ targets over the last couple of years with a 90% success rate doing what i stated.

how long has the recovery taken typically?

 

[Sutra]

What amount of exact match anchors is generally considered to be overoptimized?

 

[Bladhark]

What amount of exact match anchors is generally considered to be overoptimized?

More than 1 use. Using exact match anchors is an outdated practice and it's best to avoid the use of them at all.

Best thing to do is to create anchor texts that are variations and combinations of main keywords/topics together with a some words that describe it. So for example if a main keyword is "best dog food" you can make something like:

"read this article from BRAND NAME to discover the best dog food"
"the best dog food and the reasons why has been explained here"

Branding anchors that are surrounded by content that describes the the (main) keywords works pretty well too.

So in short it's best to avoid the use of exact and also even partial match anchors at all and never use the same anchor twice (except for branding anchors).

It's actually pretty logical if you think about it, how big are the chances someone links to a website with an exact match anchor? Excactly, it almost never happens, so keep that in mind.

 

[Sutra]

More than 1 use. Using exact match anchors is an outdated practice and it's best to avoid the use of them at all.

Best thing to do is to create anchor texts that are variations and combinations of main keywords/topics together with a some words that describe it. So for example if a main keyword is "best dog food" you can make something like:

"read this article from BRAND NAME to discover the best dog food"
"the best dog food and the reasons why has been explained here"

Branding anchors that are surrounded by content that describes the the (main) keywords works pretty well too.

So in short it's best to avoid the use of exact and also even partial match anchors at all and never use the same anchor twice (except for branding anchors).

It's actually pretty logical if you think about it, how big are the chances someone links to a website with an exact match anchor? Excactly, it almost never happens, so keep that in mind.

Dude, come on. Why even reply if you don't know what you're talking about?

 

[Bladhark]

Dude, come on. Why even reply if you don't know what you're talking about?

Oh I don't know what I'm talking about? Then explain it what's not right about what I wrote...

 

[Sutra]

It’s all good man. Go do your thing.

 

[Bladhark]

It’s all good man. Go do your thing.

I will for sure, but I'm still curious why you replied like that.

You asked the percentage of overoptiming, I gave my opinion and then you say that I don't know what I'm talking about.

I never said that exact anchors aren't working anymore, I just said that I think it's an oudated practice and there are better ways to do your anchors these days. Just because you see the whole top 10 shitting out exact match anchors doesn't mean it's a good practice :-).

Anyways, all good indeed.

 

[Sutra]

I will for sure, but I'm still curious why you replied like that.

You asked the percentage of overoptiming, I gave my opinion and then you say that I don't know what I'm talking about.

I never said that exact anchors aren't working anymore, I just said that I think it's an oudated practice and there are better ways to do your anchors these days. Just because you see the whole top 10 shitting out exact match anchors doesn't mean it's a good practice :-).

Anyways, all good indeed.

I replied like that because you stated your post as fact. However, you are wrong. You did not state your post as opinion as you claim in this last message. Had you actually stated it as opinion, I would not have replied to you at all. I would have just waited to see if someone who actually knows what they are talking about would respond with some helpful info.

You're claiming to be an SEO speciailist and posting as if you know for 100% fact what you say is true - basically trying to come off like a guru. It's bullshit that 1) wastes my time, and 2) harmful to newbies who may believe what you're saying.

 

[Bladhark]

I replied like that because you stated your post as fact. However, you are wrong. You did not state your post as opinion as you claim in this last message. Had you actually stated it as opinion, I would not have replied to you at all. I would have just waited to see if someone who actually knows what they are talking about would respond with some helpful info.

You're claiming to be an SEO speciailist and posting as if you know for 100% fact what you say is true - basically trying to come off like a guru. It's bullshit that 1) wastes my time, and 2) harmful to newbies who may believe what you're saying.

Then come with something that disproves (or at least something that debates it) it instead of coming with such BS replies.

You ask a question, I reply and then the answer is not what satisfies you or what you are looking for and then you call it "wasting your time". The only one that is wasting our time is you with these kind of replies.

Also then you are going to push me in the guru corner, wth is that? I'm not selling anything here or intent to do so. So stop your speculations please.

 

[zelch]

What amount of exact match anchors is generally considered to be overoptimized?

would be good to get back to the original question. I doubt the truthfulness of @Bladhark 's assertion that if you have more than one exact match anchor, you are overoptimized. Where is the line though? I thought I was safe but got spanked in the recent update for exact match anchors (at least I think that is what happened). I guess I will find out what % is acceptable as I seek to dilute my anchors and get my rankings back. But would be good to know what target % I'm going for where I could expect to see a recovery.

 

[Bladhark]

would be good to get back to the original question. I doubt the truthfulness of @Bladhark 's assertion that if you have more than one exact match anchor, you are overoptimized. Where is the line though? I thought I was safe but got spanked in the recent update for exact match anchors (at least I think that is what happened). I guess I will find out what % is acceptable as I seek to dilute my anchors and get my rankings back. But would be good to know what target % I'm going for where I could expect to see a recovery.

Yeah that was my bad, of course you won't get slapped directly if you use exact match anchors more then one time. But my point was/is that it's best to limit the use of exact match anchors as much as you can nowadays. The whole X amount % use of exact matches is outdated in my opinion (and experience).

To get back to your question;

The whole issue indeed with diluting now as @Ryuzaki already mentioned, is that new link update. I have to be honest that I don't know what kind of links that spam update targeted (as I wasn't affected negatively or positively myself), but it seems to be the lower quality ones (duh) that you usually could use to "fix" a link profile.

It's impossible to tell the right % of exact matches where you'll recover. I obviously don't know what the quality of links is and the amount of them that holds the exact matches. But (in my opinion) it's best that you go over each of them and decide if it's worth to keep or if you can somehow remove or disavow them.

If you didn't built too many exact matches you can also build some branded web 2.0's, but that's only worth your time if you didn't built too many of them. Some high quality (note high quality) forum links from relevant topics might do the trick as well. There are some sellers of those around.

But again, this is just my opinion and most likely what I would do, so do with it what you want

 

[mj22]

do you have a recommended provider for the niche edits or DA30 guest posts?

for the DA30's i have just been using our own stuff. For the niche edits, serpwolf or ccarter, they have some of the best deals ive seen online when it comes to cost/link.

how long has the recovery taken typically?

I have had it take 1.5 -2 months, and i have had it take 6 months. It really depends on the situation. I dont have a bunch of time to be on here, but if you would like, i would be willing to take a look at what you have going on and give you my 2 cents on it. Any info would remain 100% confidential. (dont worry, i wont try to sell you anything lol)

What amount of exact match anchors is generally considered to be overoptimized?

I was told this a long time ago, and it has served me well. 3-5% exact match + 10% partial match. Is this a one size fits all? I dont think so. I have an older site with decent authority and an established link profile, and i have gotten pretty aggressive with that site and have had no issues at all. I have client sites i have taken on that are older with good authority that also have higher than the above %'s and most(not all) of them seem to do fine as well. One of my younger sites that is about 1.5 years old that i have been building out had no exact match or partial match ran to it. A few months ago i decided to run some partial match anchors and see how it reacted. The damn thing tanked on me, it went from pos 7 for my main keyword to into the 20's. I went back to running generic and branded anchors to it and 2 months later it came back into position 8.

I think its different for each site and niche, or thats what it seems like. I dont know everything, i can only speak to what i have experienced first hand, but this is what has been working for me.

I hope this helps.

 

[Sutra]

for the DA30's i have just been using our own stuff. For the niche edits, serpwolf or ccarter, they have some of the best deals ive seen online when it comes to cost/link.



I have had it take 1.5 -2 months, and i have had it take 6 months. It really depends on the situation. I dont have a bunch of time to be on here, but if you would like, i would be willing to take a look at what you have going on and give you my 2 cents on it. Any info would remain 100% confidential. (dont worry, i wont try to sell you anything lol)



I was told this a long time ago, and it has served me well. 3-5% exact match + 10% partial match. Is this a one size fits all? I dont think so. I have an older site with decent authority and an established link profile, and i have gotten pretty aggressive with that site and have had no issues at all. I have client sites i have taken on that are older with good authority that also have higher than the above %'s and most(not all) of them seem to do fine as well. One of my younger sites that is about 1.5 years old that i have been building out had no exact match or partial match ran to it. A few months ago i decided to run some partial match anchors and see how it reacted. The damn thing tanked on me, it went from pos 7 for my main keyword to into the 20's. I went back to running generic and branded anchors to it and 2 months later it came back into position 8.

I think its different for each site and niche, or thats what it seems like. I dont know everything, i can only speak to what i have experienced first hand, but this is what has been working for me.

I hope this helps.

Yes, that is helpful. Thank you, I appreciate it.

Regarding the amounts (3-5% exact match + 10% partial match), in your experience, is that the overall distribution of links sitewide, or is that the distribution of links per page?

 

[mj22]

Regarding the amounts (3-5% exact match + 10% partial match), in your experience, is that the overall distribution of links sitewide, or is that the distribution of links per page?

On paper, yes per target url. Again it depends on the situation.

 

[harrytwatter]

1. empty anchor
2. Your Brand
3. yourbrand.com
4. Your name or pen name or authors name
5. some mishmash of above
6-infinite. anything that isn't containing the target pages money keyword

This is how I prefer to think of over-optimization, not in the specific number of anything but more in the order that the anchor ratios can be broken down into. If the order is whack then something ain't right, probably.

 

[Politico]

I'm curious to hear everyone's thoughts on internal anchor text and exact match keywords - are you keeping it low as well or using a lot more exact match anchors on internal linking?

 

[Prentzz]

Internal and external, I've seen enough evidence now that I would never use exact match anchors. I wouldn't even use partial match. I would try to use something related.

If you're trying to rank for "Seattle plumber" I would NOT use:

Seattle plumber
best Seattle plumber in the area

I would use something like:

plumbing specialists located in Seattle

I wouldn't even use:

plumbing company in Seattle

Because while it doesn't have an exact or phrase match for "Seattle plumber", I'd assume that you would also like to rank for "plumbing company".... So, don't focus so much on the specific keyword, I'd avoid including any phrase match for any keyword that you would want to rank for. In this case, "plumbing specialists" is not a popular search term, so you can sacrifice that.

I tend to be extremely cautious with anchor text, because it's something that's hard to change later.

EDIT: Adding a note here, because I know someone will read this and use only this style of anchor for 100% of their links. Don't do that.

 

[ianovici]

@zelch I've diluted links in the past with pbn links(internal network). I would do it only if the percentages are extreme, over 15%.

 

[Ryuzaki]

I agree with the sentiment, which I've also posted here before, that I wouldn't use more than one exact match anchor per page, period.

I disagree that you'll never get them naturally. I get a lot of real links with exact match anchors, which come from people with real sites who have an understanding of SEO past the pedestrian understanding. They tend to do it that way out of habit or subconscious desire for themselves, I suppose.

To the specific question of what's too much and how do we know, the industry has always thought about over-optimization in the sense of thresholds that shouldn't be crossed, and those thresholds are based on percentages. Early on, the assumption was that the thresholds were the same across all keywords and niches, and I recall at the start of Penguin when this all came crashing down, people were saying 10%. I think that was based on BoFu2u's MicrositeMasters case study, which was probably less scientific than needed and just got attention because it was the first out the gate.

As time progressed, people began getting more conservative about it, adjusting that number down to 5% and then even 3%. I'd lean in this direction too. It's not a problem I ever face because I know not to dance close to the threshold because Google will tighten up on it and screw you eventually, so this isn't personal experience I'm spitting out on hitting that threshold.

I'd honestly say to never go above 1% exact match anchor usage purposefully. It'll happen and it'll be fine. But I think that's a safe approach that's future proofed. The real question is how many instances do you need of exact match anchors before you hit diminished returns? Is one enough to get the boost out of it? Or does 100% usage give you a massive boost until you tank (it does, go look at CBD SERPs and watch them long enough to see the top winners tank after certain updates). But I feel there are also diminished returns, too.

There's enough other places to get improvement that we don't need to walk along this razor's edge.

One thing to consider is that Google will punish a tactic to get people to stop doing it, then stop punishing it when enough people stop exploiting it (because it's a decent signal they can use). But we also need to remember that Google has to keep the appearance of improvement up and justify their existence. And people will start exploiting old tactics again, and they'll get squashed again. We don't want to get caught up in that. My point is that they'll return to old exploited methods and hurt people using them again eventually, or build it into the offline calculations to regularly punish, which they do with exact match anchors.

 

[googlealchemist]

I was told this a long time ago, and it has served me well. 3-5% exact match + 10% partial match. Is this a one size fits all? I dont think so. I have an older site with decent authority and an established link profile, and i have gotten pretty aggressive with that site and have had no issues at all. I have client sites i have taken on that are older with good authority that also have higher than the above %'s and most(not all) of them seem to do fine as well. One of my younger sites that is about 1.5 years old that i have been building out had no exact match or partial match ran to it. A few months ago i decided to run some partial match anchors and see how it reacted. The damn thing tanked on me, it went from pos 7 for my main keyword to into the 20's. I went back to running generic and branded anchors to it and 2 months later it came back into position 8.

How long did you monitor things before building the generic/brand to bring it back? Any chance the partials were still just causing the typical 'google dance' before settling down higher than #7?