I make every anchor link different. That way any filters are unlikely to be tripped. Safer longer term.
Well, not sure it is 'exactly' what you want but look at this:
Actually little under 9% would be around in the middle between 8.5% and 8.9% from previous investigation.
>>I make every anchor link different. That way any filters are unlikely to be tripped.<<
You could trip the filter that says that on average some anchor text will be identical. If every anchor text is different it looks like you've gone Black Hat. ;)
|Do you take into account that a lot of sites probably disregard your anchor text request and just link to your site URL / site name? |
Yes... I look at the anchor text that other links on the linking site have gotten and make a guess about what kind of anchor text I might get.
"I make every anchor link different. That way any filters are unlikely to be tripped."
I have doubts on a filter here. I've been following one in my industry, although it's mainly a consumer resources site,yet the anchor text is the same four word phrase over and over again. The site ranks a steady #4 in google and rarely moves. Steady at #3 in Yahoo and MSN--been there for a few years.
A test site with now over 4,400 backlinks from sites that obviously see it as a good resource.
as far as I can see, every single one has used the 2 word title as the anchor.
Still on first page of G, Yahoo etc.
I personally don't pay attention to it. Penalizing for a site having the same anchor text too frequently seems a bit archaic to me. Jim's Widget Shop will likely have that over and over as his anchor text - even from completely natural links. If you think about it, there are a lot more intelligent ways "link campaigns" could be identified by the engines.
I agree, I dont think there is a filter, too many sites seem to prove otherwise :)
If you own wigets.com, and sites link to widgets.com cause its a good site, there is a good chance the site will rank well for "widgets", why would google ban this site ranking well?
Generally speaking I go after ~8 big phrases (inc singular and plurals) and many (~5-10) add a word type derivations thereof. My widget is ambiguous and the two names are interchanged and searched roughly equally - additionally sometimes it is searched for as "complete type widget" and other times "subset of widget" hence the variation
I tend to regularly use on of the 8 bigger phrases and mix an match derivations (or ocassionally lack thereof) and sometimes add pricing details - This creates for a huge amount of variation while roughly targeting each of our two word phrases and (largly incedentally) landing important three word ones.
It works for us in the sense that we are placed relativly well across the board which is very unique in our market.
>>> I tend to regularly use on of the 8 bigger phrases and mix an match derivations (or ocassionally lack thereof) and sometimes add pricing details - This creates for a huge amount of variation while roughly targeting each of our two word phrases and (largly incedentally) landing important three word ones.
Very nice post, Nalin. I practice quite similar to yours and the results are more or less fantastic.
Instead of worrying about the identical anchor texts penalties, I think it would be more fruitful in long term if you concentrate more on longer keyword phrases, similar to message #10.
The issue here is not one of penalties for using identical anchor text, it is an issue of establishing a broader semantic footprint in the topological landscape. This is not applicable to all sites, but for those sites that have many synonyms of their keywords it is better to use them all in variation in anchor text in ratio to how popular they are. In doing this it would also be wise to have some of these links point to deeper pages on the site specifically tuned to these other various keywords.
Agree with Neuron,
It makes sense, it's better for targetting deeper more focused pages and it develops a certain topology for which your site *seems* to belong to.
Also, this should help you avoid any anchor text filter perceived or otherwise.
Also agreeing with neuron and nuevojefe. See this short thread, which relates to this one, on Semantic Linking...
And, a further comment about the possible exact anchor text penalty...
Such a penalty was widely reported, but what was not reported was the nature of the sites and the source of their inbound links.
Imagine that you send out link requests and get inbounds on, say, ten of them. What's the likelihood that the anchor text on these ten links is going to be identical? Not very high, I would guess.
So, sites that had hundreds of identical inbounds might have had a measure of control over the anchor text that those of us who hunt for links might not have had. Such control suggests that other relationships among the sites might also have been noticeable... like linking patterns and server IPs, eg... and those relationships might have hurt rankings more than similarities in anchor text. But maybe anchor text is what was reported at the time. Only guessing, but this has got to be true in a large number of the cases....
My content site get lots of backlinks that i haven't asked for. Many from .gov .edu. They don't care about SEO. They just link to may homepage as "Red Widgets". No obvious penalty, the homepage has recently gone to PR6.