SEO Methods / Tips You many not already know

[takeachance]

Google has long since been in the game to idenify spun content which is where their 'ngram' patent comes into the equation. As many have said here the fact is algorythmically they are getting closer to nailing spun copies of content on a WORD level. If you spin at a paragraph AND sentence level combining that with alternate WORD variations to at least 5 sets each you are in the clear. There is no computational power equal to the task of identifying the variations which can be achieved through effective spinning.

As with all things in this world if you put in the effort you will eventually see some form of reward....

I talk from some experience that the scrapebox strategy you mention will depend on a number of different factors which are not documented here. You need to consider the typical 'link graph' associated with your niche and whether it is normal to acrue many lower quality links (look at the top 10 before you blast) such as that provided by a typical SB blast. Your recipt of incoming links needs to also be balanced with the publication of content on your site. I have run many SB experiements to money sites and can tell you that it is best to do your homework before embarking on all out war with your tools.

That is not to say, don't do as the OP suggests, I personally do research to better understand what 'normal' looks like and above all, I will take action to a varying degree which of course is the main thing - do something and not nothing

 

[d4rkterror]

thanx gonna give some of this a try

 

[_Chip_]

Wow in site anchor is new to me.... Cool... I will start using them in my articles ...repped

 

[ForeverNever]

Gonna refer back to this one...

 

[Cindy]

Thanks for the info! A couple of questions for you:

If I already have a 700 word article with the keyword in it a total of 3 times, would you create another occurrence of the keyword for the anchor link, or use one of the existing 3 occurrences of the keyword? Basically I'm wondering how these anchor links affect your keyword ratio.

Are these anchor links good for ranking in general or just to get sitelinks?

 

[lancis]

Here are some ways Google is likely trying to fight spun content:

Google {does not|doesn't} yet {know|know for sure| know clearly} that this is {spun|randomized|random}, but the sentence {structure|placement of words} is consistently in the same place.

TBS has a list of synonyms that can be randomized and automatically spun. Google is very likely trying to reverse engineer this, and "unspin" them.

Alternatively, Google may try using different synomyms for different sentences and see if there are exact match.

Also, in spun content ,the anchored term is likely similar or the same, or the URL is likely the same from each paragraph.

_______________________
Summarized: Google doesn't yet KNOW how to find this, but it is their every day effort of 1000's of engineers to find a way to get rid of spun content.

My point is that it works now, but may not work in the future.

Its comments like this that make me wonder whether SEO people do their homework properly...

Google doesn't try to "unspin" anything, Google performs statistical analysis, which can come in a variety of ways: starting from simple "hist" and ending with more advanced kernel density estimation.

You can rearrange all the words in the article - its PDF (probability density function) wouldn't change. In the same way you can spin the words - its PDF will change only a little.

And if you have a database of synonyms/translations and perform LSA (http://en.wikipedia.org/wiki/Latent_semantic_analysis) on the article even translated or word level spun article will come out as identical to the original one.

 

[gogoman]

Thanks .........

 

[jerdude]

Thanks for sharing your research! Sometimes I xrumer to my sites too. some instances it takes months to get over the dance and rank and sometimes it doesn't at all. Hmm...

 

[thetermy]

Thanks for the info! A couple of questions for you:

If I already have a 700 word article with the keyword in it a total of 3 times, would you create another occurrence of the keyword for the anchor link, or use one of the existing 3 occurrences of the keyword? Basically I'm wondering how these anchor links affect your keyword ratio.

Are these anchor links good for ranking in general or just to get sitelinks?

Firstly, keyword density is a guideline not a science.

Secondly, does the keyword show up enough in the article to show relevance? I'd say you should put the keyword a few more times in the article for a total of 7 times, but up to 10-14. Also, ensure that you include a couple variations, not just the exact keyword;

Red tie shoes

red shoes with tie

etc.

 

[thetermy]

Its comments like this that make me wonder whether SEO people do their homework properly...

Google doesn't try to "unspin" anything, Google performs statistical analysis, which can come in a variety of ways: starting from simple "hist" and ending with more advanced kernel density estimation.

You can rearrange all the words in the article - its PDF (probability density function) wouldn't change. In the same way you can spin the words - its PDF will change only a little.

And if you have a database of synonyms/translations and perform LSA (http://en.wikipedia.org/wiki/Latent_semantic_analysis) on the article even translated or word level spun article will come out as identical to the original one.

Interesting.

Are you saying that the result is the same; Google is able to / close to understanding spun content?

 

[pietpatat]

In point 1 you say scrapebox to your $page can't hurt you, else competitors would do this to you.

In point 3 you says spun content will hurt you, while competitors can also do this to you.

Doesn't make any sense...

 

[lancis]

Interesting.

Are you saying that the result is the same; Google is able to / close to understanding spun content?

Considering the number of NLP (Natural Language Processing) specialists they have, I would say that content spun on the word level is easily figured out by Google.

From my experience with NLP its much harder to figure whether article is duplicate or not when you replace entire paragraphs and sentences. However, even here there is a catch.

There are algorithms today that are able to summarize the content automatically, i.e. they take 1000 words article and make a very short summary out of it.

So if you have 1000 spun articles, their summary might be still the same in spite the replaced paragraphs/sentences. Note, that summary in this case is not human readable - its a mathematical entity used in NLP. So its pretty robust and can be used to filter out duplicate content.

 

[WizIMS]

Wow man :

AWESOME POST!

Respect.

 

[cyrilchua]

That is very interesting to me.. I do indeed believe google has the capability to read out spun content of the word level on the net. But I kind of doubt that with that nmuch bull on the internet all having spun content, google can manage to do such analysis on all of them.

 

[verilix]

This is great, except using Scrapebox on pages that already have thousands of comments, pointing to your main URL isn't the best thing to do. I'm not saying it'll negatively effect you, but there's always that chance.

 

[thetermy]

Considering the number of NLP (Natural Language Processing) specialists they have, I would say that content spun on the word level is easily figured out by Google.

From my experience with NLP its much harder to figure whether article is duplicate or not when you replace entire paragraphs and sentences. However, even here there is a catch.

There are algorithms today that are able to summarize the content automatically, i.e. they take 1000 words article and make a very short summary out of it.

So if you have 1000 spun articles, their summary might be still the same in spite the replaced paragraphs/sentences. Note, that summary in this case is not human readable - its a mathematical entity used in NLP. So its pretty robust and can be used to filter out duplicate content.

I follow. What is your conclusion?

Are you in favor of "cleverly" spun content ie. paragraph redistribution

 

[thetermy]

Wow man :

AWESOME POST!

Respect.

Made my day

 

[oneder]

That is very interesting to me.. I do indeed believe google has the capability to read out spun content of the word level on the net. But I kind of doubt that with that nmuch bull on the internet all having spun content, google can manage to do such analysis on all of them.

Seriously, I've read plenty of articles on the net written by different people copying each other that say the same thing over and over, and in summary form would be dupes?!?

 

[thetermy]

Seriously, I've read plenty of articles on the net written by different people copying each other that say the same thing over and over, and in summary form would be dupes?!?

Yea, this bugs me too. But Google needs to come up with something to understand regurgitation. Maybe they devalue recycled content once there have been x amount of posts on it?

 

[Chicilikit]

I have been thinking about this "HERE" and "LINK HERE" in anchor texts, but I always read from seo guy that I should use my keywords as anchor text and change it just little not so much, but now I'm going to change the way I use anchor texts. Thanks for this