Keyword distribution in the whole site

[DavideM]

I've been taught during a SEO course that the whole site has to contain the chosen keywords with a fixed proportion of optimized pages, that should be like this:

50% of pages optimized on the most relevant keyword (just one keyword)

25% of pages optimized on secondary kewords (depending on the size of the site, could be a few pages for each secondary keywords)

25% of pages on long tail keywords.

the teachers was a very respected SEO professional, but I've never seen this strategy anywhere in other articles or SEO guides.

what do you think about it?
It's true that it brings visibility for the top keyword?
does it lead to cannibalization?
what others strategy do you use?

[kadesmith]

My definition of relevant in this case is: the keyword is often used in sentences with the term.

If my keyword was cold cereal, the words milk, bowl, and spoon would be relevant.  I could then build pages for the long tail search terms:

Cold cereal bowl

how much milk should I put in a cold cereal

What is the best spoon to use with cold cereal

On each of those pages, I would link to my main page which is optimized for the keyword cold cereal using the anchor text "cold cereal", but only if it made sense to me as a user to see that phrase linked.  Sometimes you have to be creative in your content copy, but most of the time it can be done and make sense to the end user.

Google is very good at recognizing these relevant keyword patterns.

[DavideM]

It does make sense. But what is your exact definition of "relevant"? How is the keyword used in the content?

[kadesmith]

It is important you keep in mind the terms relevancy verses optimized.  A page can be relevant to a topic (main keyword) without being optimized for that keyword.  When this is the case, you can focus on a long-tail keyword for a sub-page and have relevancy for the main keyword.   You would then want to link to the main page with the main keyword from the sub-page.

In this case I would say that what you learned in your class is true, keep AT LEAST 50% of the pages RELEVANT to the main keyword, but not optimized to the main keyword.

The end grading factor though is, what works for your site?  A lot of SEO is trial and error based upon bsaic principles.  Sometimes you just have to try what your gut tells you to do and watch and see if that works, if not try the next idea.

[DavideM]

it was a one day beginners course, and we was talking about a 100 pages site targeted to national audience (italy) as an example.

but was presented as a general startegy.

Now I'm not a total newbie anymore, I have worked on some sites with goods results using this  technique.

Nowadays I'm working on a 100 pages site, targeting to USA whole market. I've find relevant keywords for the market and I'm making decision about pairing pages and keywords.

My question is: is the strategy illustrated good for a site like that? how many pages I have to optimize for each relevant keyword? I have about 30 keywords, and 8 among them  are the most importants

[Bryan_Loconto]

Hi David,

Build your site for your audience, create good content they will want to read, learn from or simply be entertained by. For on page SEO use the seomoz on page optimization tool and target one keyword per page. You can also use scheema.org to add some meta descriptions, this will helo the search engine determine what you page is about.

For external link building:

50% Anchor branded name / URL

25% Diverse anchor text

25% Exact Match

Hope this helps.

[kadesmith]

The first thing I would take in to consideration is, how much did you pay for the class?  Was it a quick one day thing, a couple hours long or a week?  How in depth was the instructor able to go?  For a complete newbie just getting in to SEO with a small website (4-5 pages) these would be some good tips because it would give them experience optimizing pages and watching a keyword start to climb.  With a site that small, it's most likely something that is trying to climb in a local search.  In a small populated area doing local SEO, this would work very well to rank in most areas.

There is a possibility this could lead to cannibalization, but it all depends on how you structure the 50% of pages that are "optimized" for the keyword.  Ask yourself, "Is there one page that is clearly about my keyword that others reference?"

Other strategy to ask yourself is, "Does it make sense to the user or just the search engine?"  Search engines themselves don't buy anything.