Deep linked content

[leporello]

I have a question regarding deeplinked content (hope that is the correct term).

Say I have a rakeback site, but one of my tabs is Poker Equipment where I sell poker chips and poker tables. On my poker tables page I am trying to rank for the term "Poker Tables". Will my poker table page be ranked based on my sites overall search engine ranking? For example if I rake number 3 for "rakeback" will I naturally rank highly for "poker table" even though this is on a different section of my site.

On a similar note, how important is it to link build for the page www.blahblahblah.com/PokerTables in order to increase my SE visibility for the search terms Im targeting or is the overall value of my site naturally passed on to the rest of the site.

Thanks for any advice/input and I hope that my questions are clear!

[Scott]

I'm no professional by any means, but I would suggest building links to your poker tables page directly for the keywords you're trying to target on that page.

As for your second question I think that your PR value is the only thing that passes through your pages pending there linked together properly. This will help your deep pages become more autoritive I would presume, but I don't see it helping you rank for the keywords on that page unless you've been building links for the keywords to that page.

Often when targeting longtail keywords you will be able to rank with just the content and without any links to the page due to the low competition, but I'm sure for "Poker Tables" you'll need some nice inlinks pointing to your poker tables page.

Not sure how much of this is correct, but I'm sure others will weigh in and correct anything that is wrong.

[happyjack]

It can't hurt to rank well for one, word, but it may not help a word that has nothing to do with each other. You need to sculpt it a bit.

[Michael Martinez]

Link building is the least efficient method for expanding your search visibility.

Think of search visibility as the measure of how many queries for which a given page is found, modified by how high in the rankings for those queries the page appears.

Search engines take hundreds of factors into consideration when determining rankings (Google alone claims over 200 factors determine rankings).

In a hyperoptimized query, all the leading sites are usually trying to rank through anchor text -- in other words, they have invested heavily in competing through links. The bonus comes from the repetition of your keywords, not from the PageRank.

I have occasionally see pages outrank link-rich pages simply by repeating keywords on-page often enough to overwhelm the search ranking algorithm's weighting mechanism. This often happens on blogs where hundreds of comments are embedded on a single HTML page.

[Hazo]

I have heard that repeating keywords and keyword density is a myth on several occassions. Are you saying the opposite?

[Randy Ray]

Repeating keywords is not a myth.

Keyword density should be greater than 0%, but other than that, I wouldn't worry about density in terms of a percentage. (It would be impossible to implement any kind of keyword repetition on-page without having a keyword density of more than 0%.)

[Michael Martinez]

Keyword density is a metric that was developed for document indexing systems in the 1960s or thereabout. It was adapted to Web search indexing by second-generation search engines in the mid- to late-1990s. It was devalued or ignored by third-generation search engines like Google, Inktomi (Yahoo!), Ask, Microsoft/Live, et. al.

Repetition is a basic element evaluated by IR (Information Retrieval) scoring algorithms (of which there are far more than I could ever hope to enumerate).

All major Web search engines decide their rankings on the basis of two scores: an IR (document content evaluation) and LV (link valuation) score.

In Google's case, their IR scoring mechanism is partially described on their Web site. They look at things like font sizes, emphasis (bold, italics, etc.), repetition, and so on.

Their LV mechanism includes (internal) PageRank, which is based on an extremely complex filtering mechanism (most links probably do not pass PageRank these days). Another aspect of Google's LV mechanism is anchor text.

Google treats inbound link anchor text as if it is a part of the document being linked to. Hence, there are many examples of documents where query terms only appear in links pointing to the documents.

The documents achieve high IR scores through repetition and maybe emphasis (no one has yet devised a test, of which I am aware, to show that bolding and italicizing link anchor text helps the destination document) that is bestowed through link anchor text.

Generally speaking, if you can repeat a word 100 times on the page (without tripping any spam filters) you stand a pretty good chance of ranking well (highly competitive queries may require hundreds of repetitions).

Because old-school search spammers used to repeat keywords hundreds of times in meaningless text jumbles and text lists, most SEOs have abandoned the use of repetition in their on-page copy, fearing they might trip spam filters. However, I have found that as long as the repetition occurs in coherent natural language (it should make sense to a human being reading it), the search engines usually accept it without any problem.

In the old days Google simply added the (internal) PageRank score to its IR score to adjust every document's rankings. Now they appear to be using PageRank in more sophisticated ways. For example, documents that lack sufficient (internal) PageRank are stored in the Supplemental Pages Index (which Google started hiding in 2007). Supplemental Index pages are not permitted to rank first if there are less relevant, higher-PageRank documents that still meet the query's basic requirements.

It may be that Supplemental pages are only partially indexed by Google, or maybe hashed (their text represented by complex numerical strings). Hence, it's desirable to obtain at least some PageRank for as many pages on a site as possible, so that:

• They are allowed to compete for high rankings
• Their text is fully indexed
• Their links are allowed to pass value (anchor text, PageRank, trust, etc.) to other pages

That's probably a little more information than you were looking for.

[Hazo]

Thanks a lot for the detailed explanation Michael! Cleared some things up for me, especially about keyword repititon.

[leporello]

Thanks Michael, fantastic in depth reply. Much appreciated.

[Andrew Scherer]

I noticed a big newbie mistake is pointing all your links to your home page.

About 90% of the links I build go to deep links/subpages.