How often do you link without anchor text?

So a while back I had a site get penalized, pretty much disappeared from Google. Pretty sure it was because of excessive anchor text (used the same anchor text every time I linked).

This time around I'm being very careful, and I always alternate between like 10 anchor texts as my main keywords.

Now I have another concern. Is it bad to always link with anchor text (even if you alternate the anchor text)? How often should you link without anchor like this:


I'm worried because I've barely been doing that, and I was thinking of ordering some profile to link to my site without anchor text to kind of make things look more natural.

Good idea?

Have you ever had a look at sites that rank well in your niche?
In my observations, I have seen sites ranking #1 for keywords that are used as anchor in less than 10% of their links.
This is one of the first things I do when I start a new site, establish the brand.
If you have worked with SEO clients, you can see how beneficial this is.

Example.

A client of mine had about 400 links when I started her campaign. All the anchors were her company name or just the url.
She ranked nowhere for any of the 4 keywords we were targeting. I have made only about 30 links with keywords as anchors, and now she is ranking in top 5 for all 4 keywords.

Many may not agree with me, but I firmly believe in establishing your "brand" before you get into using keywords as anchors. I just think it looks more natural.

themidiman said:

Have you ever had a look at sites that rank well in your niche?
In my observations, I have seen sites ranking #1 for keywords that are used as anchor in less than 10% of their links.
This is one of the first things I do when I start a new site, establish the brand.
If you have worked with SEO clients, you can see how beneficial this is.

Example.

A client of mine had about 400 links when I started her campaign. All the anchors were her company name or just the url.
She ranked nowhere for any of the 4 keywords we were targeting. I have made only about 30 links with keywords as anchors, and now she is ranking in top 5 for all 4 keywords.

Many may not agree with me, but I firmly believe in establishing your "brand" before you get into using keywords as anchors. I just think it looks more natural.

Click to expand...

Thanks for going in depth! I guess I should stop focusing so much on specific keywords and more towards branding, and powering up my sites backlink profile.

If anybody else has a different view on this, I'm all ears!

I have numerous top rankings from using the same anchor text. I'm talking about 30-40 sites.

I too have ranked alot of sites(well over 100) using anchortext.Think about a url that doesn't have keywords in it,it needs good anchortext so google knows what the site is trying to rank for.
In fact all the initial links I do on any site are anchortext so I cement the keywords I'm ranking for.Once the site starts ranking for the keywords I'll vary the anchortext but not til it's in the top 2 pages.
I have seen sites ranked very high for keywords the site has nothing to do with,this happens when you don't anchortext properly and google decides what you're trying to rank for,I am talking the content and site description had nothing to do with the keywords it was ranking for.
You do need variable anchortext but I work 70% keywords 30% varied without keyword(click here,site url,names).On high PR links it's not usually possible to get your desired anchortext anyway.
Of course an EMD is telling google what you want to rank for but I still use the keywords as anchortext for the initial links.Another reason is let's say someone wants to try and bugger your site by using bad anchortext.If you have firmly established keyword anchors it is highly unlikely to even move your site,however if you have alot of varied anchortext you're in trouble.I've tested this on my own sites,sites with too much variable anchortext will start bouncing all over the place a site with a high % of relevant keyword anchors won't move unless the links are constant over days and days but even then you can simply counter by doing more keyword links.

We've done some serious testing on this over the last 5 years; and contrary to what a lot of people will tell you, it's very important to have a wide range of anchor texts...

We always ensure there are a percentage of brand and URL anchors (both www. And without) plus company/site name etc. also you need a good wide selection of generic anchors ( the 'click here' type of thing). Then you need a wide variety of phrase-matched anchors, so for 'weight loss' you have 'weight loss considered', 'understanding weight loss', 'weight loss secrets' etc. basically lots of different long tails in effect, but with your phrase always inside in the correct order (phrase matched.)

I know there are loads of people on here saying that you can just blast 10k links to a page with only 1-3 anchor variations; but for most sites this is rubbish. If you've done loads of link building and have established a good background 'noise' of other varied links, and your site isn't a new one, then you can often get away with this, but as a progressive and long term strategy it's defunct with the advent of Panda.

I've tested and monitored this across several million links built via all the main link types, and all other things considered, this is one of the largest causes of over optimisation on newer sites: keyword anchor > keyword domain> keyword title > keyworded content. It's just too much in Google's eyes.

If you think about it logically, and consider how you would write Googles code, then it's obvious that this is going to be seen as manipulation...

Or ask from the other side; how would it look if all your links DID come organically and naturally. What kind of anchor profile would there be. You'll then find that my original statement is obviously the case.

Our Link Coder tool does all this automatically as a matter of course, and is one of the reasons we created it; so our VA's and link builders cannot mess with our SEO. It expands out the keyword distribution curve and adds domain/generic anchors as well as extended automatic long tails and phrase-matches. We also use this on our Backlink Banzai linking program, and have been for 3 years now.

This is one of the most important changes in SEO, and together with title over optimisation on your link properties, is the main reason for newbie early failure in my tests (apart from the obvious not doing enough work!)

FYI we SEO for thousands of sites, and have in the region of 500-600 of our own, and so we do have the volume for sufficient empirical testing.

on my private network i use

60% non KW related (click , info ....)
15% KW with additional words ( best KW, TOP KW, information about KW)
25% KW exact

since my network is pretty new i have no clue yet if that is good or not .

abrigado said:

on my private network i use

60% non KW related (click , info ....)
15% KW with additional words ( best KW, TOP KW, information about KW)
25% KW exact

since my network is pretty new i have no clue yet if that is good or not .

Click to expand...

You could probably increase the percentage of phrases here if you wanted, but you're being cautious, which is sensible.

As a rule of thumb, do 1/3 generic and domain anchors, 1/3 mixed exact match, and 1/3 mixed phrase match/long tail. This way you end up with 2/3 using some connotation of your keyword phrases, but they're highly mixed and random.

Our generic phrase list is over 200 long, and our tool expands keywords out by over hundred-fold. Variety is the key to looking natural. Very few organic links look the same, and google would expect a good third to be nothing to do with your phrase/subject.

What about high competition niches where when you analyze the top10 competition for a keyword, and none of them have URL backlinks or non-keyword backlinks... In fact, they're almost exclusively "widgets" "widget reviews" etc

I'm definitely in agreement with jascoken, but I just wanted to see what others think regardless. I'm not sure there is a definite answer to this. But like I said, a while back, I had a site disappear from Google. I was building links gradually, so I know it wasn't me building links too fast. The problem had to be that I only used ONE anchor text.

I'm vary cautious now. I don't have a specific split like you mentioned above (1/3, etc..), but I do use quite a few variations. For my home page, I use around 10 variations. For my inner pages, I use 3 or 4 variations.

FuryFit said:

What about high competition niches where when you analyze the top10 competition for a keyword, and none of them have URL backlinks or non-keyword backlinks... In fact, they're almost exclusively "widgets" "widget reviews" etc

Click to expand...

Are you sure? I think in high competition niches, the top 10 sites naturally get tons of links from people, so for them, it's not much of an issue building links with specific anchor text. It's being balanced out with the natural links coming in from others.

The upper people said is very good and it's right to vary the anchor text?

As general think, google know the site is consist by many content .
so if you have a few anchor text ,they will think your site is rubbish.what's more,it's can attract spider to crawl your site

thedon23 said:

Are you sure? I think in high competition niches, the top 10 sites naturally get tons of links from people, so for them, it's not much of an issue building links with specific anchor text. It's being balanced out with the natural links coming in from others.

Click to expand...

Yes, definitely - but maybe I'm just in spammy niches... Before I sold it I had a site ranking in the top 3 for phentermine so my perspective must be skewed haha...

I guess if you go type in "weight loss" you'll find sites with a more natural link profile.

FuryFit said:

What about high competition niches where when you analyze the top10 competition for a keyword, and none of them have URL backlinks or non-keyword backlinks... In fact, they're almost exclusively "widgets" "widget reviews" etc

Click to expand...

There are many different scenarios where this 'can' happen, but without looking at a specific link profile I can't tell you. I can tell you that the law of averages says that for every one that does work, there will be 10-100 that didn't work in this way. And SEO is often about playing the odds, if you want to be around long-term.

There are so many ways to 'game' things short term, and if there were not that many links ad they all came from high authority sites, then you might get away with this.

I've NEVER seen a long term successful link profile that was purely exact phrase anchoring. If Panda has showed us anything, it's that a longer term strategy is now more important than ever. How many people got burned this year? I haven't had a single site delisted or knocked back in the rankings through all panda updates. And we Backlink hard, believe me.

At the end of the day, there will be hundred of people on here who can say, 'but I did this - and it worked really well!' I work with many clients who we've achieved superb results for over the last 18 months, but the methods we used wouldn't work now. The landscape is changing, and so are the effective methods.

You don't have to take my word for it. Get out there and try it on 100 domains and see what your stats are. When we test, we don't do it on 1 domain, we do it on hundreds.

There are always exceptions to the rule!

jascoken said:

There are many different scenarios where this 'can' happen, but without looking at a specific link profile I can't tell you. I can tell you that the law of averages says that for every one that does work, there will be 10-100 that didn't work in this way. And SEO is often about playing the odds, if you want to be around long-term.

There are so many ways to 'game' things short term, and if there were not that many links ad they all came from high authority sites, then you might get away with this.

I've NEVER seen a long term successful link profile that was purely exact phrase anchoring. If Panda has showed us anything, it's that a longer term strategy is now more important than ever. How many people got burned this year? I haven't had a single site delisted or knocked back in the rankings through all panda updates. And we Backlink hard, believe me.

At the end of the day, there will be hundred of people on here who can say, 'but I did this - and it worked really well!' I work with many clients who we've achieved superb results for over the last 18 months, but the methods we used wouldn't work now. The landscape is changing, and so are the effective methods.

You don't have to take my word for it. Get out there and try it on 100 domains and see what your stats are. When we test, we don't do it on 1 domain, we do it on hundreds.

There are always exceptions to the rule!

Click to expand...

Well, I didn't backlink hard, my link profiles aren't as varied as yours, and on all my current projects, I have seen sideways movement and in some cases been bitten by panda.

So, I definitely will be upping my percentage of non keyword and non anchor links. I've always been surprised how few of my competitors even do brand anchors.... They're 99% keyword anchors with little variation.

Great post.

Does this work for target pages or do you guys think this works for the site in whole? LIke if you have multiple pages you are building links to and each page you target one to two keywords. Is this over optimizing? I assume it is

Dan Da Man said:

Does this work for target pages or do you guys think this works for the site in whole? LIke if you have multiple pages you are building links to and each page you target one to two keywords. Is this over optimizing? I assume it is

Click to expand...

Yeah, it would be. I think 1 or 2 anchors wouldn't be enough. For example, if I had this page:

widgets.com/blue-widgets

I'd personally link with the following anchors:

- blue widgets
- cool blue widgets
- widgets blue
- navy blue widgets
- widgets that are blue
- etc..

thedon23 said:

Yeah, it would be. I think 1 or 2 anchors wouldn't be enough. For example, if I had this page:

widgets.com/blue-widgets

I'd personally link with the following anchors:

- blue widgets
- cool blue widgets
- widgets blue
- navy blue widgets
- widgets that are blue
- etc..

Click to expand...

Exactly!

But really try to get numbers up. I talk to 'professional' SEO's regularly who come up with 20 anchor variations and think they're done...

As an example, If I built 1,000 links, then I'd seriously want 50-67% of them to 'contain' my phrase, but I'd want them all different if possible. Can you imagine how natural this looks !?

I have a public page of 'generic' variations if anyone wants to use them:

http://www.backlinkbanzai.com/GenericPhrases.htm

Dan Da Man said:

Does this work for target pages or do you guys think this works for the site in whole? LIke if you have multiple pages you are building links to and each page you target one to two keywords. Is this over optimizing? I assume it is

Click to expand...

Dan

This definitely also applies 'deep', not just to home pages. I create home-page URL anchors to deep pages as well, but I don't usually use the entire URL (unless it's short) as you can end up really long anchors otherwise.

Also, using URL anchors can often add to your keyword backlinking too don't forget if the URL contains your keywords in the file name...

jascoken said:

Dan

This definitely also applies 'deep', not just to home pages. I create home-page URL anchors to deep pages as well, but I don't usually use the entire URL (unless it's short) as you can end up really long anchors otherwise.

Also, using URL anchors can often add to your keyword backlinking too don't forget if the URL contains your keywords in the file name...

Click to expand...

Do you mean:

if url is http://www.site.com/widget-topics/widget-training.php

you would link to it using:

http://www.site.com
and
widget-training.php

FuryFit said:

Do you mean:

if url is http://www.site.com/widget-topics/widget-training.php

you would link to it using:

http://www.site.com
and
widget-training.php

Click to expand...

Yes!

With that URL ( under 40-50 chars) I may also use the whole URL as well. But you know what some Wordpress and 2.0 URLs can be like... 100-150 chars is not uncommon.

I always ask myself, what would I do if i were a webmaster of this site? Or if i was Google looking in. This normally gives you the correct answer with some thought and reasoning.