This is one of Kurt Melvin's BIG PAGES. Also see:
Big Page of SEO
|| Big Page of Tuelz Tips || Big Page of Free SEO Stuff

The Ultimate Guide to Modern On-The-Page
Search Engine Optimization for Multi-Domain Networks

     

What makes this page different from all the other SEO pages on the Net?  First, this guide focuses on "on the page" SEO. While "off the page" is also vastly important, it seems that it dominates today's discussions and "on the page" has been forgotten.

Next, this page assumes you are performing SEO and site upkeep on a lot of pages and domains, while the other SEO guides seem to deal with search engine optimization a page at a time. Optimizing a single page, or even a single domain, is far different than optimizing across multiple domains, a necessity in the modern SEO world.

Finally, I've created a companion page, Kurt's Big Page of Tuelz Tips, that gives step by step details showing exactly which techniques to use for optimization and how to implement strategic changes to improve your pages' potential to get free search engine traffic to their maximum potential. We don't just say "what" to do, we also show you "how" to do it.

This page is a multi-media experience. If you own any of the Tuelz listed on this page and your Tuelz.exe is running, you can just click on the links and you will be taken to the corresponding Tuel. Note: You must start the Tuelz program before clicking on any Tuelz links on this page.

Even if you don't have Tuelz, this page is still tremendously informative and the resources and Kurt's Big Page of Tuelz Tips will also explainhow to optimize pages in-depth.
 

Is Reverse Engineering and Research
Wasting Your Valuable Time?

If you've researched anything about SEO for any length of time, you've probably seen software, services and articles about studying search engine results for things like "keyword density" and other common traits.

I disagree with this to a large degree. I'm  know I'm going against "common wisdom", and I'm not suggesting you should never research the Search Engine Results Pages (SERPs). What I am saying is, your focus should be on being productive by making pages and domains.

Why I disagree with most SEO advice concerning researching SERPs:

Where to Put Keywords
In reality, almost every word on your page is a "keyword", with the potential to bring traffic from the SE's. And here's a secret, you really don't chose the keywords for your pages, people using search engines do. We hope our keywords match the words users type better than the other guys' pages.

For our discussion, we'll define keywords as those words we are targeting for their "value" and concentrating on them for strategic SEO purposes.

Using the tips and strategies outlined on this Big Page of SEO, you won't need to worry about exact formulas and densities. Instead, you'll be learning a time-tested strategy to build pages using as many different "formulas" as possible, with the minimal amount of effort.



When optimizing many pages across multiple domains, your goal is NOT to find the "perfect formula" for SEO, but rather to use a mixture of as many different SEO formulas as reasonably possible.

Tracking and reverse engineering things like "keyword density" is useless. You will learn how to create pages that contain a wide variety of keyword densities, so if the engines change you'll likely already have a page with a favorable keyword density.


In the Big Page of Tuelz Tips you'll learn how to create sites and pages that account for different "recipes" for SEO in a natural way by mixing and matching various elements of SEO outlined throughout this page.

This page is NOT a tutorial on how to perfectly optimize a single web page. Instead, it is how to optimize thousands and thousands of pages (or more) using the widest variety of SEO elements and keywords. Just as the top results for any decent keyword don't follow an exact recipe, neither should your sites and pages.

If the engines change, as they often do, then by varying your strategies on multiple pages you will often have other pages that meet the new algorithm's preferences, adding a lot more potential to all your SEO/SEM efforts. Your keywords should be used in vast number of ways, numbers and densities, and throughout as many pages and sites as possible.

Keywords can be placed in many places on a page, including:
 

The Most Important  Places for "Basic" On-Site SEO

Domain Name:
URL/File Names:
Page Title:
Anchor Text linking to the page:

www.Keyword.com www.Key-Word.com
keyword.com/cookbooks/recipes.html
Keyword One, Keyword Two
Your KEYWORD Here
=> page with keyword in title tag.

If you learn nothing else, know that the four above are probably the most important ways to use keywords when performing on-site SEO.

However, this doesn't mean every domain name and every URL you use should contain keywords. It simply means you should FAVOR putting keywords in domains and URLs, doing so more than you don't, as there are plenty of high ranking sites that don't have keywords in domain name or URLs.

Here's some more ideas for where to place keywords:

Meta Tags - Not very effective any more, if at all. However the description meta tag is still used to describe your site in some engines, and the engines can always change. While meta tags probably aren't/won't be of much use for ranking, the description tag may be of value.

On pages you do use meta tags, remember these rules:

  1. Make them unique from page to page
  2. Don't "stuff" keywords, meaning no repeating keywords or using irrelevant keywords
  3. Try to make sure any keywords used in tags are also contained in body text.
  4. You can try adding a few typos to the keyword meta tag. If the meta keyword tag has any value, it may be with misspelled words.

Remember, the meta tags have little, if any, importance to your ranking. However, you may still want to include them, especially the description tag.

Suggested Usage:

This is a good blend intended to cover any changes the engines may have, as well as act as a decent test for you to track the importance of meta tags into the future.


H1, H2 and H3 Tags

<h1>keywords</keywords></h1>
 

Font Attributes such as italics, underline, size, color.

<font color="red">keywords</font>  <big>keywords</big>
 

Embedded File Names, such as: keyword.css , keyword.cgi , etc.

Instead of:
style.css

Try:
keyword.css
 

In ALT Text:

<IMG SRC="picture.gif" ALT="keywords" BORDER="0">

Note: If you combine the two immediately above, you could have the code look like this:

<IMG SRC="dog.gif" ALT="obedience training" BORDER="0">
 

In Email Addresses:

keywords@YourEmail.com

This is another variation of "anchor text linking". Of course, you need to worry about spam email address harvesters, as well as real people may send you an email....However, having keywords in an email address on your page is an option to try to improve SEO ranking.
 

In CGI Input Forms:

When possible, you can add keywords to CGI boxes, such as in the examples above. This may be beneficial to use with search boxes, etc, where the keyword is already filled out for the visitors.
 

HTML Comments Tags

<!-- Here's a comment tag with Keyword, another keyword and another keyword. -->

Comments tags can be inserted into the HTML code and are NOT displayed on the web page in a browser. They are used so the web master can make notes to him/her self.

WARNING: Due to abuse, it's likely stuffing keywords in comment tags will do more harm than good. However, it is something to keep your eye on, test on a very small percentage of pages, and use the knowledge when examining the SEO of other web pages.

In addition, some of these can be combined, such as making a link in bold tags and /or in header tags:

More Apple Pie Recipes and Articles

Note the keyword phrase "apple pie recipes" is actually contained in three different tags:

  1. An <h3>  </h3> tag
  2. anchor text
  3. and italics.

  

Other Possible Factors for Your Keywords:

Keyword Density, includes in body text, anchor text and various html tags.
Density is simply a ratio comparing the number of times a word/phrase is used on a page (or in a tag), compared to the total number of words on the page.

We don't spend much time on this concept, as the strategies we will learn on this page will take into account densities and our goal is to use a wide variety of keyword densities, so issues like this take care of themselves, without intensive research.
 

Proximity is the term used for multi-word searches and tells us "how close" the keywords are to each other. Basically, if the person doing the search enters the keywords and they are right next to each other , then it will usually be best if your pages also have them next to each other.

For example, if the search was blueberry pie, then these two words should be next to each other, and would be considered more relevant than a sentence such as:

"We went to the store and bought some blueberry jam. We we got home we decided to eat a chicken pot pie."

You can see in the example above that the PROXIMITY between the two words in red is 12 words, and wouldn't be as relevant to a search query for "blueberry pie" as a page that had the two words next to each other (in close proximity).
 

Order of the Words
The order of the keywords is also very important. Since people often search by typing keywords in different orders, it is optimal that you also use valuable keywords in as many different order as possible.
 

Prominence
Is the keyword at the beginning of the page and/or tags? The general wisdom is early and late on the page, as well as in title tags.
 

Example of KEYWORD with HIGH Prominence in tags: Example of KEYWORD with LOW Prominence in <title> tag:
<tittle>KEYWORD, Welcome to our site!</title>
<h3>KEYWORD, Welcome to our site!</h3>
<tittle>Welcome to our KEYWORD site!</title>
<h3>Welcome to our KEYWORD site!</h3>

Note that when KEYWORD is listed early in the tag it has high prominence.
This concept also applies to the <body> text (content) as well.

Give most of your keywords high prominence in your pages and tags. But not all the time. Since you have multiple pages and multiple domains, mix it up having some pages without using KEYWORDS at the beginning of your content and your tags.

Rough Guideline:
50% use High Prominence.
20% Middle
20% Late
10% None


Clustering - Clustering is the theory that elements of a page are related, such as links "clustered" together on the bottom of a page are probably navigation links. It is a good strategy to include some of your strategic links from "non-clustered" links, such as linking a word or phrase in a middle of a sentence to an important resource:
 

Example of Clustered links: Example of Non-Clustered Links:
Apple Pie
Blueberry Pie
Cherry Pie
We went to the cafe for lunch and I had apple pie for dessert. I usually skip lunch but was craving something sweet and had the turkey and swiss sandwich, coffee and salad, with a slice of cherry pie for dessert last time, if my memory is correct.

Misc. SEO Factors

Size of Page Files - Have most of your pages 5-20k in size, but don't be afraid to get some bigger pages, from 30-100k, maybe 2-10% of your pages. You may even want a few over 100K, just as a test.

Using Good HTML Code - There's reports that having good, valid html could help your rankings.

How Big, How Many Pages Does Site Have? It seems Google prefers larger sites. However, you should mix up the size of your domains. Some as small as 5-10 pages, others with 10,000 pages, or more.

How Mature is Domain/Site - Older, well established sites seem to do better in Google than newer sites.

Length of Domain Name Registration - Some theorize that registering your domain name for multiple years can aid in your site's ranking. Makes some sense, as it increases the "price of entry" which would eliminate spam, but finding concrete evidence either way would be difficult to prove. However, if you can afford to register a domain name for 3-4 years, why not? It probably doesn't hurt and at least you don't have to worry about renewing it for a while.

Local/Regional Info - If your pages concern a geographical area, then having (or not having) identifying information relating to your sitel's location may affect your rankings.

Examples of Regional Information that may affect ranking:

This includes on your pages as well as your domain name whois.

If you can, on websites about regional topics sometimes mix in contact info, in  atypcial business format, here's one example:
 street address
 city, state, country zip
 555 555-5555

Warning: Don't use fake phone numbers. If they belong to someone and they get harrasses, it your fault and be prepared for some backlash.

Page Load Speed - The load speed of a page may be a factor for a couple of reasons.

First, fast loading pages are beneficial to the visitors, so it could be used as a ranking factor.

But even if page speed isn't directly included in an algorithm, there is a practical application: Almost assuredly the engines use some time of "time out" or length of time they spend on any given site or page.

So, it stands to reason that if  two sites each have 100 links on them, and Site A loads twist as fast as Site B, that Site A has a much better chance of getting it's pages indexed each time a spider visits. Having said this, we can't assume that the engines treat all sites and pages the same. A well respected site may get more time for each spider's visit.


 



Be sure to make site map pages and all strategic pages load as quickly as possible
for maximum "spiderability".

Use your Moniterz to make sure your pages are up and responding quickly to
ensure your pages are spidered to their fullest.



 

Can Your Pages Pass the Fresh Test?
The engines, particularly Google, like pages that update and stay fresh. Here's some questions you need to answer:

Not every page on every domain of your's needs to update every 5 minutes. Spread it out and mix it up. On some of your domains, simply have 10 pages, with a couple that update a few times a month. Others, you'll want to update a few times a day, across plenty of pages.
 
 



Important Note: Newbies are inclined to over optimize. There is no such thing as a perfectly optimized page that ranks highly for a keyword of value.

Instead of applying every strategy on this page on each of your pages, mix and match these concepts throughout your network of sites.



 

Links and Anchor Text
Anchor text, both internal and external KeywordAnchor=>Keyword.html=><title>keyword

To hyphen or not to hyphen file names? While traditionally it was recommended to use hyphens, this is also a strategy that may have been over-used by SEOers and spammers.

We recommend that you mix it up from one domain to another. Try something like:

Getting Pages Spidered
Getting your pages spidered is the first step of SEO and is essential. "Spidered" simply means a SE visited your page. If it meets the standards of the SE and is ready to be included in the SERPs, it is called being "indexed", meaning your pages are in the SE's index.

Links act as the pathway the SE spiders will use to find all your pages. The best way to send the spiders to your new pages is to get a link to them from a page that  you know is indexed and already in the engines. This becomes easier as you build your own network of sites.

In theory, the most direct way to ensure all your pages on your site are within two clicks of each other. In other words, if you visit any page on your site should be able to get to any other page on your site within just two click.

To do this, you must:

For smaller sites, your homepage can and should, act as your domain's site map, linking to all pages, with all pages linking to it.

For larger sites: Use multiple site maps, have links to these site maps from your homepage. Link all pages on domain to your homepage.

However, site maps also have weaknesses, one of which is that your competitors can easily see all the pages on your domain. Site maps can also contain too many links and exceed SE limits, and may well be "over optimized" by having a high density of anchor text containing similar keywords and phrases.

Still, site maps are a vital aspect of SEO and getting pages spidered and indexed. However, below we show you some alternative techniques that can help get pages spidered, as well as help your optimization for increased traffic potential.

Once your pages are included in the index and the more often you update your pages, the more often search engines will usually spider your pages. If you notice any pages drawing the spiders consistently, use these pages as your site map and hallway pages, adding links to your new pages to get them indexed. Whenever possible, let the engines determine what pages of your's should be site maps.
 

Use Respected and High Ranking Sites to Improve Your SEO
It is possible, and logical, that you "inherit"  a little of the content, relevancy and reputation of the sites you link to. If linking to "bad neighborhoods" is bad, then we may assume that linking to good sites is a good thing, at least to the point of having some of our pages link to high quality sites. 



Use FatSearch to find sites and pages that rank highly in a variety of major engines. Since these sites rank highly in multiple search engines, there is a very good chance these sites are the highest quality and keyword relevancy.  

Note: This requires that you have downloaded and installed the free tuelz, and tuelz.exe is running on your PC.

Then, "cluster" your own highest quality links with a few of these high quality, highly relevant links, for SEO as well as branding. This is a good strategy for SEO as well as for your site's visitors.



 

More Advanced Tuelz Linking Strategies:

Linking pages in a sequential manner can also be beneficial. For example, you've probably seen a series of pages that are linked in this manner:

      <= Back || Next =>

This strategy can be very helpful from a usability standpoint, aqs it allows your readers to follow a progression of links.

From an SEO point of view this effectively gives all pages included at least TWO links from other pages, as putting two links to two other pages assures that each page also has to links pointing back to it. This not only aids in the spidering of your pages, it also eliminates ORPHAN and DEAD END pages. See below "AVOID: Orphans and Dead End Pages" for more info.

Another Tuelz strategy we call "Related Navigation", which is the same as "also see", where the links suggest more and additional pages with similar content. 

For example, if you have a page about "dog food", you may want to link to all of your other pages that include the phrase "dog food". This is has two tremendous benefits:

  1. Gives our users links to our other pages that offer more content on a similar subject.
  2. Linking to other pages that also use the same word phrase can increase relevancy for BOTH pages.

 
 
People Rank: How Humans May Influence Rankings
People Rank is what I call the concept of how real people can affect the ranking of webpages. While much of this isn't really "on the page", the content you have on your pages can directly influence the following factors.

Engines such as Google can track many activities through their own server logs, cookies, Google Analytics or from the millions of people that use the Google Toolbar. They can they gather this information and apply it to other components of their algorithms..


About the time I wrote this article, Google was worth about $120 BILLION dollars, so let's take a look at the finances of hiring human reviewers...

Hire 1000 people at $10 an hour, 40 hours a week. That's $400,00 a week and they only need to look after the top 10-20% that probably brings in the largest portion of their ad revenue.

In other words, it would cost about $1 a month per keyword for the million most profitable keywords. That's not very much money. If you owned Google, wouldn't you pay $1 to make sure your most important keyword results were top notch.

You can play with these numbers in a variety of ways, we just used these as an example to show how efficient this could be. And again, there's been a little evidence leaked to back up this "conspiracy theory".
 

A Theoretical Algorithm Using People Rank

A little preface...Many modern SEOers believe there are 3 elements to Google ranking, in order of importance:

1. Hilltop and TrustRank

2. External Link Pop, Link Relevancy, PageRank

3. On the Page

Let's theorize for a moment and visualize ourselves as Mr. Google. How would you rank web pages?

Here's how a "People Rank" system may work:
I might have a ranking criteria based similar to how dog racing and horse races have "classes".

When starting their racing careers, dogs and/or horses are placed in a lower "class", usually a qualifying race. If a dog/horse wins a race, it moves up a class and if it loses, it drops a class, eventually seeking its level of competition based on actual head to head contests.

And if it keeps winning it will keep moving up in class until it gets to race for the big bucks in Class A.

In the case of a search engine, a web page on page 5 would compete with the other sites on page 5, and move up or down dependent on the page's performance versus the other pages listed.

Let's say Google's "SandBox" is really just a "qualifying race", where NEW pages containing certain keywords were given artificially high rankings when first indexed to establish high enough click through numbers to make tracking human behavior on the pages accurate.

("Sandbox" is a term used to describe the way Google seems to rank a new page highly for a couple of weeks, then the page disappears.)

The artificially high initial rankings may be needed to give enough clicks to a page to have a decent sample to use for "people rank", which will be recalculated after the "people rank" data has been collected.

Google could use some of the more sophisticated tracking techniques to measure:

...And include the results as part of "People Rank" in their relevancy formulas.

It's easy for them to do, either via their toolbar or the use of cookies. I contend that using actual visitor surfing habits is a far more accurate way to determine "quality", than links, PR, Hilltop, meta tags or other SEO. Although, it doesn't have to be one or the other, as any engine can use "All the above", and probably should.

If a page gets a higher "people rating", it moves up a "class" (in this case, a higher search engine results page) to compete against better competition, and if it gets a lower rating, it moves down, or is dropped.

I doubt this is the exact process they use. And it's not intended to be. Instead, this is meant to get you thinking about the possibilities.

While I am most probably wrong about the exact process, I do contend that using "people ranking" is valid for SE ranking, I believe this method adds a degree of secrecy to the stats. Much of the info the search engines use to rank pages is "publicly available"...It isn't that hard to do keyword density research, or see if the keyword is in the title tags. external linking pages, etc.

Using "people rank" not only can help provide more relevant results, it also is a security measure to protect the other elements of the algo, since they would be impossible to account for using available numbers. The data collected from "people ranking" would be private and protected and adding these variables, unknown to SEOers.

IMO, the engines will begin to rely more and more on real people ranking factors, especially for the most profitable keywords...It's seems that using human behavior and rating not only makes sense, it also seems to make cents.
 

Trust Rank
Trust Rank is how Google determines the value of a link. While Trust Rank is a little off topic and not "on the page", I'll mention it here, as it also affects how you plan on inter-linking multiple domains.

How well respected are the sites that link to you?

Every link to your pages doesn't have the same value. For example, links from domains using .gov or .edu were thought to be more highly respected by Google than links from other extensions. Google's logic was that links from .gov and .edu aren't available to everyone and therefore have more "trust".

(This is further blurred by the possibility that SEOers have exploited this, so the value of .edu and .gov links may not be as powerful as they were recently.)

Other sites may also be "whitelisted" and thus their links carry more weight, such as respected sites like cnet.com, Yahoo's paid directory, DMOZ/ODP, etc.

Other factors that could determine "Trust" that are "on site":

FatSearch: Trust Rank
 

Hilltop
Hilltop is a complex system believed to be used by Google. Without going into the gory details, understand the concept that the relationship between two sites can influence how much impact linking between the two sites has.

For example, the links between two sites that are known to be "affiliated" won't/shouldn't count as much as a link from a totally "unaffiliated" site.

"Affiliation" in this case is simply any biased relation ship. This relationship can be determined by your domain name whois, IP addresses, etc.

FatSearch: Introduction to Hilltop
 

The You Rank - Should All Those Free Google Tools Scare YOU to Death???

This section is based on speculation and theory. The only "proof" I have of You-Rank is based on the common-sense principle that this would be what I would do, if I owned Google and had access to the information they are known to have.

The You Rank is my prediction of an additional
method Google either is, or will use for ranking purposes.

In the most basic terms, the You Rank is based on what Google knows about YOU, and then applies this information to any and all of your domains and pages.

Consider, there's now article spinners that can and will create 10,000 doops of a single article, a "billion blogs a second" applications, PLR duplicate sites, link exchange clubs that artificially inflate link count and more, all of which are intended to "beat" the engines.

In the past, the SE's have used "on the page" criteria such as the words on your webpages, pages titles, etc. They also factored in "off the page" criteria such as who links to you.

Think of YOU RANK much like a credit report. If you have "good" sites in your past, any site you create in your future may get preferential ranking. If you've created spam sites or used other techniques Google doesn't like, then your next sites and pages may not be as well ranked, based on nothing but YOU.

Be Afraid....Very, Very afraid. Everything Google Wants to Know About You, and Doesn't Bother Asking

Please read the following, add it all up, then really think about what it all means. This may be the most frightening stuff you'll ever read on the Internet...

To be perfectly honest, Google is no longer a search engine "company", they are a data mining company, looking to extract as much information about YOU as they possibly can...They are doing this to target advertising to you, based on who you are, what your interests, what's your businesses or employment, plus things like your sex, age, marital status and lifestyle.

If you use any of the following "free tools", think about what you are letting Google know about you, and how they may apply it to all your sites.

Toolbar - Do you have the Google tool bar? Google can easily track every site you go to and store that info. But it doesn't stop there. If you notice, the Google bar also has a spell-check feature to use when posting on forums, blogs, etc. This means Google can even track everything you've ever posted on any blog, forum, discussion group, anything you've ever typed into a web form, including passwords, etc.

GMail - It's fairly well known Google indexes and tracks your email in order to target advertising on a contextual/keyword based method. In other words, if you use GMail, Google knows everything you and those you've emailed have written. Think about this...They even index the content of email sent to GMail accounts, even though those people haven't accepted the Gmail TOS.

Gmail even auto saves DRAFTS when you don't want it to. You may trash the email you were going to send but did Google?

"UnSocial" Networking? Plus, Gmail users must be sent "invitations" before they can register for accounts, so now Google can cross-link you with everyone that you sent an invitation to,  as well as the person that sent you your first invite. Now, they don't only know you, they know about the company you keep.

Desktop Search - Have you downloaded the Google desktop search engine that scans your hard drive so you can search your own PC? Now Google knows what's in every web page or text file that's on your hard drive.

Google Analytics/Web Analytics - Now Google knows everything there is to know about your websites. How much traffic you get, where you get it, your keywords and more. Use Analytics and Google knows more about your web business than you do.

AdSense/AdWords - If you use either of these, Google knows who you are, where you live, what your bank account is, your Social Security number...And more...Plus, each and every webpage you use with your AdSense ID tells Google EXACTLY who you are, no more info is needed.

Google is a domain name register - You may not even be able to hide your Whois info...While it is doubtful  as to whether Google has access to your hidden Whois, it does give Google direct access to all the domain name records that are public, allowing them to store and access historical data concerning whois records.

Affiliate Programs - If you've ever put an affiliate number of your's on the same page as AdSense, then Google can easily and accurately cross-reference these numbers and tell exactly who you are.

For example, if you used Amazon and AdSense on one domain, the only Amazon on another domain, Google still knows EXACTLY who you are, where you live, what your bank account number is, etc. Remember, you gave Google this info when you signed up for AdSense.

Then, Google can cross-reference any other affiliate numbers they find on your pages and see if they match an other fingerprints they have detected of your's.

Google Checkout - Google's new entry into payment processing similar to Paypal, but now they also know exactly how may sales you've made and to whom.

Google Apps for Your Domain - This new service is meant to compete with Microsoft's business applications. But the Google programs are free and online.

But there's just one catch: The Google Apps privacy policy says that Google will give you a "domain manager", a real person that essentially has access to every bit of information entered, including your email and any data created by Google Apps.

Google Bugging Your Home? - Get this, according to TheRegister.co.uk .... If you have a microphone plugged into your computer, Google is even developing a way of spying on the background sounds made in the privacy of your own home. Google is attempting to detect what TV programs you watch and the music you listen to, but claim they aren't interested in personal conversations.

Are you scared yet?

If you use the programs above, Google can scan and store all your emails, all info on your hard drive, stuff you type into input forms on all the websites you visit, read your list of website bookmarks, read the data created from Google online applications, track your searches, listen to your TV and stereo, track all your web activity, see how many sales you make, analyze your server logs and access stats better than you, cross reference them with AdSense and AdWords info to tell exactly who you are...Even your GMail trash may not be safe....That is some serious info!

 

The You Rank Revisited
Now what do you think about my theory of "YOU RANK"? Everything I listed above records something about your behavior, and if you are using a few or more of these Google services, Google probably knows more about your Net habits that you do!

How would Google tie this in with the SE ranking? If Google has all this info about you and had banned a few of your sites as "spam", they can easily include this data as part of any or all of your other domains. If you have any affiliate number they can cross-reference with any of their other info, that alone could be enough to get your pages banned, in this theoretical situation.

In other words with Google tracking your every move, if you are a known spammer to them, your domains could be dropped simply based on who you are, not what's on your pages. On the other hand, if you have high quality sites, that make a lot of sales and get a lot of traffic from links, Google may give your pages a boost in the rankings, possibly based on who YOU are.

Google has spent a lot of time, effort and money developing programs that collect data about all of your computer activities.

To repeat, Google really isn't in the search results business any more, it is in the data harvesting business with a goal to collect as much data about you as possible, for marketing purposes.

Whether or not Google uses any type of "You Rank", hasn't been established as far as I am aware at this time. However, it is well established Google is collecting as much information about as many people as it can.

The question now becomes: Since they have all this info, why not use it for SERPs? Why not?
 

Footprints, Fingerprints and "DNA"
These three terms relate to the "clues" you leave that identify you with a website, or even two of your websites together.

Here's a non-inclusive list of "fingerprints" the engines can use to determine if your sites are related:

I believe Google spends a lot of their SE resources trying to determine "relationships" between two sites to check if the linking is legit. Remember, Google became famous because they use linkage in their algo rankings, and to protect this, they must be able to determine if any given linkage to a domain is legit or SEO spam.

If Google detects any similar "DNA" indicating two sites are "related", the effect of the linkage will most likely be diminished.

 

Gold - Silver - Tin - Protect Your Assets
Here's a BIG mistake I see many noobs (and others) making...You create 100 sites all at once, then just as fast as you can, you slap AdSense on them.

This can be "Suicide by Google"...The instant you put AdSense on a website, Google knows exactly who owns the site, what your name is, where you live, etc, and maybe even your bank, if you use direct deposit.

Remember, you told Google all this info in your AdSense account, and all they have to do is associate your AdSense ID number with your account info. You may as well just have a big "kick me" sign taped on your back...

Think in terms of "Value Levels", as well as the old "80-20" rule, which theorizes that 20% of your domains will bring in 80% of your profit.

Web Site Value Levels:

  1. Gold Level Sites - The Top 20% of your domains. These are established sites, already have a steady stream of quality traffic. With your Gold sites, don't play any games and use only high quality, original content and make them purely "whitehat". These are your sites that use AdSense and other high-quality revenue streams that leave fingerprints.
     
  2. Silver Level Sites - The Next 20% of your domains. Silver sites fall somewhere between Gold and Tin. Use other affiliate programs for these. Maybe take a few chances with these, but not a lot. These are your sites that get SOME traffic, but aren't your "best" sites.

    Use Silver sites to build links and drive traffic to Gold sites, as well as capture follow up info.
     
  3. Tin Level Sites - The majority of your sites should be Tin. These are your "experimental" sites that use methods that may be of higher risk, such as scraping, cloaking, SEO spam, etc. The purpose of these sites is to experiment, as well as point links and drive traffic to your higher value sites.

    On these sites, you don't want anything that can ID you, such as same whois or IPs, or even affiliate numbers for ANY program that Google can match. Use the Tin sites to push traffic and create links to your legit Gold and Silver sites.

Remember, if you have 100 domains, 20 of them will bring in 80% of your profits (an example). Save your AdSense and other premium revenue streams for the 20 you know are making money, not those you "hope" will make money.

Once/if a Tin site starts getting traffic, then work to "legitimize" it and make it a Silver site, then promoting your best Silver sites to Gold Level status.

Don't be in such a panic that you reveal all your cards too soon by slapping your AdSense on all your sites the instant you launch...

It is OK to have sites with NO AFFILIATION on them, as a matter of fact, it's recommended. Remember the Google patent application that refers to "thin affiliates"? What if you don't have ANY affiliate/order buttons? Hmmmmm...Kind of throws a wrench into the "thin affiliate" theory, now doesn't it?

Have some patience and maintain your privacy from Google as long as you can, until you are positive a site will generate some decent profit.

(Note: Don't hold me to the 80/20 as being exact. Your results will vary. Just be more patient and smarter than the other gal/guy.)
 

Reference Chart Depicting Various Linking Strategies to Help Reduce Fingerprints:

 

To the right is a combination of theoretical linking strategies.  

The one point that should be taken from this chart is to mix things up and don't use a single strategy.

Even if one strategy is better than another, I have a feeling using it over and over will make it the worst strategy.

Also, there's definitely other linking strategies and variations and mixtures...

First, the graphic probably needs some updates, so don't take it as "written in stone".

Needed updates to graphic:

-The LINEAR - OPEN CIRCUIT shouldn't have each site link to the hub, but only a couple.

- The "Orange" group needs arrows from corner to corner, so all sites link with all others.

The "safest" linking strategies are:

  • LINEAR - OPEN CIRCUIT
  • FOCUSED

And you should probably use these two type strategies the most often.

You don't have to create this network in a week. You just need to outline it and plan ahead, preparing your hub, then setting out to create mini nets that point to your major hub at your own pace.

[Image]
Each color group represents a keyword theme and each dot equals a completely unique "id", including whois, IP address, as well as content mix and general "style".

Summary:

Color = Keyword theme - Each color should be a very tight niche, with each dot an even smaller niche, based on keywords and theme.

Dot = Unique ID, including Whois info, IP address, DNS, no affliate program fingerprints shared, etc.

In theory, we're trying to create an "authority site", the kind of site that you would kill to get a link from.

Instead, we're hoping to have that site for ourselves. To be honest, you won't be able to generate "authority status" from self linking, but this linking chart is still an important concpet to understand and utilize.

Hub sites should be heavy on text and lite on links, especially external links. If you have original content and articles, this is the place to put it.

Use your hub(s) to sparingly and strategically link where linking is most needed.

I think the biggest mistake I see is that most of you go after 10,000 keywords on one website/domain/ip/whois.

Instead, you need to saturate your niche from every angle, instead of spreading out into a billion niches.

Let's define "degrees of nicheness" as niches, mini niche and micro niche.

Each group of colors should be a collection of mini and micro niches. For example, the GREEN group could be a group about the mini niche "dogs", with each dot representing a domain about: dog breeds, dog food recipes, obedience training, etc.

These domains then point to the relevant pages on your main hub, in this example using the "FOCUSED" linking strategy.

Basically, I may want to create a "mini hub" for "dogs", point all the micro niche sites such as "dog food recipes" at it, then point the mini hub at your major hub.



A good niche isn't the one with the best/most powerful keyword. It's the one with the MOST GOOD KEYWORDS that you can spin off in micro and mini sites, then point at a main Gold Site.


Artificial Intelligence and Relevancy
The search engines are constantly trying to find ways to improve how to determine what web sites and pages are really about. Using computer programs to understand human speech is a tough proposition, but something we need to understand to forecast the direction the engines are probably headed.
 

Themes
The engines are trying to develop the ability to determine the theme of a page, by examining the other words that are also on the page.

For example, a page about the "San Diego Chargers" (American football team) and "battery chargers" should contain different words. "San Diego Chargers" web pages will have words like "touchdown" "passing" and player names, while a site about "battery chargers" will use words like "ac/dc", "12 volts", etc.

This flies in the face of traditional SEO advice that says to focus on one main keyword phrase. I disagree with the traditional advice, and always have. In my experience, you are far better off using as many related keywords and phrases throughout your pages and sites.

Again, virtually every word on your pages impact how the engines see your pages. And, every word and more importantly, every combination of words can impact your ranking in the engines. But even more importantly, every combination of keywords brings another possibility that your pages will match more queries.

If the engines are using "themes", then a strategy of using as many related keywords and phrases, across as many pages and sites as possible, becomes even more likely to be the most effective strategy. To me, it never made any sense to not use other focused keywords on pages for SEO/SEM goals. The principle of "themes" should make obsolete the other advice that says to only target one "keyword" per page.
 

Word Stemming
This has been around for a few years, and engines turn stemming "on and off" from time to time. When I wrote this, Google has started using stemming.

Stemming is simply using all grammatical variations of a word and treating them as similar: Clean, cleaner, cleanest, cleaning would all be treated as the same words, assuming the engine had stemming "turned on"
 

Latent Semantic Indexing (LSI) also known as Latent semantic analysis (LSA)
LSI is a process that attempts to determine the "theme" of a page by studying how often words are repeated in a text, then applying a complex mathematics formula to the words to determine relevancy.

In addition, LSI attempts to find the theme of a page by looking at, and comparing the search query to other words on the page.

For example, a search query for "apple" could easily bring up legitimate sites, relevant to the following "themes". By reading other words on each page, can you determine a different theme for each? If so, you now understand LSI.

Main "keyword": Other words on the page used by LSI to determine theme:
Apple mac, osx, computers, software, ipod
Apple records, album, john, ringo, paul, george
Apple orange, peach, fruit, produce, pie

As a thinking human, you can easiliy understand that in the three examples above, the word "apple" applies to a totally different THEME. And this is what LSI attempts to do, use other words to help determing the theme of a page.

In order to increase your on the page relevancy, you will want to use words Google thinks are related, according to their LSI, and use them in combinations through out your SEO and content.

To determine which words Google believes develope a theme for keywords, just do this simple search...Add the ~ tilde, (squiggly line thing) to the begining of your search, like this:

Google LSI Search for:
~music

Now, look for the words highlighted in bold. You should see words like: radio, sound, mp3, musical, audio and a few more marked in bold.

Tips: This LSI search only works for "major" keywords.
Also, set your Google preferences to return 100 results, so you can see a lot more pages easily.

You will want to work any LSI related keywords into your content in natural sentences and paragraphs, as well as include them prominently in your SEO efforts, anchor text, domain names, file names, urls, etc

In general, you will want to use the LSI words as your "root" for your other keywords, where the root word is part of muliptle word keyword phrases, such as "cheap ipods" where "ipods" is the root keyword.

More Info: FatSearch Latent Semantic Indexing
 

Multiple Algorithms?
Since we know there are three major engines, Google, Yahoo! and MSN, we know there are at least three different algorithms that can bring our pages significant traffic.

But why do we assume each engine only uses one ranking "formula"? Why couldn't Google use more than one?

There's actually some pretty strong evidence that Google does use different algorithms for different types of searches. For popular, one word searches, Google seems to rely more on "off the page" stuff like links, PR, etc. But for the longer, more obscure searches Google relies more on a algorithm that favors "on the page".

These longer, one of a kind searches shouldn't be ignored, as it has been speculated by many SEO experts (including me) that these type of searches may account for 40-60% of all search traffic.

Also, each engine has also been known to change their algos, and will very likely do so again in the near future. Because of this alone, all SEOers must deal with the concept that they need to plan for many different types of algos.

But let's take it even farther and discuss the possibility that Google and the other engines may actually use a combination of algorithms for a good part of the SERPs.

Casino slot machines rely on what is knows as a "random number generator", or RNG. This RNG is their absolute insurance that no one can break the sequences, since there are no sequences.

This very same principle could easily be used by the SE's. After all, slot machines are simply computers designed to win money and use computer programs, just like any old search engine.

If I ran a search engine, I'd want to protect my algorithm at all costs. If this algorithm ever became known, it would greatly endanger my results and could put a BILLION dollar company in danger.

Because of this, I'd incorporate some random factors. There's a million ways to do this.

For example, with "if/then" statements:

If the page title contains 4 words, then the body should have keyword density of 6%.

If the page title contains 6 words, then the body should have keyword density of 20%.

If you factor in the other 100+ elements that Google claims to use, you can see how they could easily create algos that couldn't be figured out.

SEO's would need to analyze an infinite number of combinations. And "infinite is impossible".

Or, how about this theory. Let's assume Google has experimented with a number of algorithms over the years and have found 3-4 (or more) that work differently, but all return good results.

Why not randomly apply different "formulas" to different pages?

When a page is submitted or found, the engines could easily apply a RNG to that page to determine which algorithms is used.

So, in theory a top 10 results page could actually be displaying a multitude of different algorithms.

And to make it worse, these algorithms are being decided by an RNG. Therefore, you will never see a sequence in the search engines results.

For example, the SERP for "German Shepard", the top 10 results may use an RNG sequence of: On the other hand, the SERP for "Poodle" may have these random algorithms:

2
4
3
3
5
6
1
6
2
2

5
5
3
7
8
4
6
2
2
0

*Each number represents a different ranking algo/formula that the SE has determined valid by testing.

So, "Poodle" may assume it's one formula that ranks higher, while "German Shepard" could say it's totally different.

It's important to remember that I'm not saying that Google "randomly" assigns "relevancy". I'm saying they could easily randomly assign different "good" algorithm that they have researched and tested.

No algorithm is perfect. But I believe a combination of algorithms would provide a wider variety of results and probably better results. It always seemed that if the engines used the exact same algorithms for every page, that pages in the SERPs would look a lot more alike.

But, we see a wide variety of "seo strategies" amid the top SERPs. Each of the Top 10 had different keywords densities, proximities, et al.

We could even take the random algorithm a step further and apply it to each of the 100 elements Google claims to use.

For example:

Is keyword use in page title?

They could repeat this throughout the page for all elements.

I don't see much difference between a casino using RNG to protect their results and a major search engine using RNG to protect their results. If anyone could "decode" either, it would be financial disaster for both.

While I doubt they are using the exact system I'm using for this conversation, I find it very likely that the engines use RNG in some degree. Maybe not exactly the same as my examples, as there's an endless variety of ways RNG could be used.

A while back, I posted a question on a forum about which of the following factors would return the most relevant results:

  1. On the Page
  2. Pagerank - External Links

Each of us could make good arguments for either side and both side were debated, resulting in good reasoning for using either/or, or both.

But in reality, Google doesn't have to choose. It could use a blend of both, or many algorithms. I've always contended that a blend of sites, some with high PR and some with high on the page factors, as a group of results, would be more relevant to more people than simply using one or the other.

Google doesn't return a single link in its SERPs. It returns 10 or more, if the user has opted to have more results returned in their preferences.

If we assume that both PR and "on the page" have merit, and each could return useful results, then why not assume Google uses three algorithms:

  1. Relies heavily on PR.
  2. Relies heavily on "on the page".
  3. Weighs both of these factors evenly.

That's three algorithms.

But each could have variations:

1. Relies heavily on PR.

A. If title has keyword, then body has X keyword density. Give boost.

B. If keyword is repeated in anchor text, give boost.

This creates 3 algorithms for the "PR" algorithm:

  1. 1
  2. 1A
  3. 1B

Three algorithms blending PR and On the Page could be manipulated in a number of ways.

Google could test millions of the variations to find 10-100 variations that return the best results. By using these "best results" algorithms in random fashion, I believe they could return very relevant results, as well as maintain their security.

They could even "spike" the results a little. For example, they could use algorithm 49-F twice as often as algorithm 07-Q.

Still don't buy my random theory?

Here's what we know as fact: Google, and all engines, make changes. And since we don't know what these changes will be, or when they will occur, "randomness" plays a large role in SEM, and unless you know what these changes are and when they will take place, then randomness plays a large part of SEO/SEM.

We've heard of "Austin" and "Florida" changes in Google rankings. Did any of the major SEO "experts" predict these changes before they were made, therefore these changes really become "random" changes. I

I'm sure Google made strategic, logical changes. But from our point of view, Google's changes were totally random, since we had no knowledge of when they would happen or what the changes would be.

After any of the SE changes, all the SEO experts will scramble to debate and "figure out" what went on. By the time anyone gets a grasp on things, makes the changes, then waits for re-indexing....What are the odds Google will make another "random" change in the meantime?

Again, SE's index words. They will always index words. I suggest concentrating on the things the SE's can't change. Like putting as many unique combinations of words on as many pages as possible, using a variety of "flavors" (styles) and keywords...



Search engine algo's will change. Always have. Always will. Playing a game of reverse engineering after each and every change only guarantees a life-long game of cat and mouse.

By implementing the strategies detailed on this page and in the Big Page of Tuelz Tips , you won't need to worry about densities and such stuff, as you'll be making pages that cover all reasonable possibilities. Instead of researching what the engines want, you should be productive making pages using a wide variety of SE friendly "formulas" in anticipation of changes.


Bad Stuff You Should AVOID

AVOID: Cuss Words, Hate Words and Slurs
While some may think it's cool to throw the frequent f-bomb in their blogs for the Rant-O-The-Day, it's likely this form of speech is hurting their SE rankings. Remember, the sites returned in any SERPs are a reflection of the search engines themselves and they have billions and billions of pages to choose from, why select a page that's full of hate and bad language?

AVOID: Poison Words
Poison words are words that are rumored to hurt your ranking if you use them. The follow list is just an example and is NOT etched in stone:

Examples of potential "poison": words:
about, awesome, BBS, best, bookmarks, casino , click, contact, cool, credit, credit cards, directory , eBay, fresh, gambling, guestbook, help, home, homepage, largest, links, mortgage, new, policy, privacy, resources, search , search engine, sex, top, TOS, updated, viagra, website

AVOID: Bad Grammar and Poor Spelling
If Microsoft can have a grammar check in Word, then they can run a grammar check on pages in their MSN index. And f they can do it, so can Google.

I first heard this rumor back in '97-98 or so, so this is a theory that has been around for quite a while, and it makes a lot of sense. Well-written and well-spelled pages probably do correlate with higher quality information.

AVOID: Splash Pages with Little Text
Search engines index words, always have, always will. Unless you have a bunch of external links pointing at your page, it is essential that you have enough text for the engines to determine relevancy.

AVOID: Thin Affiliate Pages
A "thin affiliate" is defined in a Google patent application as a web page or site that has little or no purpose of existing, other than as an affiliate of another company.

Much SE spam utilizes affiliate programs. I suggest that you not add any revenue streams or other fingerprints until a site has established traffic.

Again, if you have 10 domains, only the top 1-3 of them should have any decent revenue streams. Avoid the temptation to put revenue steams on every site you have, especially before they even have any decent traffic. Be patient and don't play your calls until you have to.



Tip: Not having affiliate programs or other revenue streams may mean some tin sites are non-commercial. This allows more liberal uses of many services, such as scripts, content, images, RSS pheeds, etc. Be aware of resources that may now be available for you to use to build sites.


 

AVOID: Over-Optimization Penalty (OOP)
I mentioned this earlier and it needs repeating...Newbies are inclined to believe "more is better" when it comes to SEO. However, in the last few years a new theory has been brought forth and it is that the engines will penalize pages that use "too much" SEO.

And this does make sense as the pages that use a high degree SEO techniques are often spammy, and not of the highest quality.

So, avoid thinking you must stuff keywords every where to get good rankings, when in fact, the opposite may be true. Instead, learn all the potential places that keywords can be included and use this knowledge to examine high ranking sites and their use of keywords.

And use this knowledge to mix a variety of SEO formulas.

AVOID: Poor Linking Hygiene
As pointed out above, be careful about who and what you link to...Stay away from "link farms" and exchanges where links are exchanged in masses.

Some reciprocal links are ok, just don't over-do it...In addition, offer "triangular" link trades, where THEY link to your Site #1, but you link to THEIR site from your Site #2.

AVOID: Redirects
Whenever possible, don't use flash redirects, frames or refresh tags that send your pages' visitors to a different web page. If you absolutely must redirect pages, be sure to use the 301 Redirect and not another method.

FatSearch: How To Do A 301 Redirect

AVOID Simple Deception:
Don't try to hide text by using the same color as a background. Also, AVOID stuffing keywords in meta tags and other areas of the page.

An old SEO trick was to use a single, "invisible" graphic to hide links to other pages. This was done so that spiders would follow the links, but humans wouldn't see the graphics...This practice should also be avoided, especially when there are better ways to hide links such as just using bigger graphics. For example, you probably have other legit graphics that can be used for discrete/covert linking.

AVOID: Drawing the Attention of Your Competition
More than likely it won't be the SEs "catching" you for something, as there's a very good chance it may be one of your competitors "ratting you off" by reporting you to the engines. Again, this is another form of "People Rank", as most engines offer a means to report "offending" sites so you should always consider that every site you do may have to pass the inspection of a real human being.

AVOID: Extensive Self- Cross Linking
We've pointed out many of the ways Google tries to determine if the two sites sharing the linking process are "related" in any way. Through Hilltop, Trust Rank and Fingerprints, Google tries very hard to check if links are from legit external sources, or if the links are "self generated".

If you are going to "self link", be sure to avoid any fingerprints for maximum effect.

AVOID:  Orphans and Dead End Pages
Orphan pages are pages that don't have any other pages linking to them, and Dead End pages are pages that don't contain any links out.

Make sure every page you want in the SEs has at least one link to it, and one link on the page pointing out.

AVOID: Stealing Copyrighted Material
Reports are Google responds forcefully if presented with documentation that a site has stolen content from another site, in addition to any legal penalties you could also get your sites banned.

AVOID: Unrealistic Updates:
Above, I mentioned that "fresh" content may help your rankings and spiderability. However, too much change too soon may hurt your pages.

A common ploy is to use random content elements that change each time a page is visited to simulate updating. This can still be a valid strategy in some instances, however, it does leave fingerprints. If a spider visits twice within just a second or two, and each time the content changes, this is a big tip-off that the page contains random stuff, and is not updating naturally.

Just because updates are generally a good thing doesn't mean every page you have must update every 10 seconds. Instead, use reasonable time frames for your updates and new pages for best affect.

AVOID: Issues with Dynamic Pages
Don't use too many attributes, parameters, variables or session files in URLs.

Any URL that contains a ? is "dynamic" in nature and can give spiders fits. Make sure the URL doesn't use "sessionID=" or a lot of characters after the ? .

If you know how to use htaccess, SSI or mod_rewrite, you can use your server to make these pages appear to be "normal" html pages, avoiding any of the possible hassles of dynamic URLs, while reaping the benefits of dynamically created pages.

One example use would be for a forum like Simple Machines. On its main index pages, it uses session IDs. Simply call this page using SSI (and .htaccess) to "pull" this page into a regular html page, then link to this html page. The engines can now follow all the links to the individual forum threads for maximum index-ability.

FatSearch: Intro to SSI and .htaccess

AVOID: Fancy Programing Stuff
Java, javascript, flash, frames are all things to avoid, if SEO is your top priority. While I'm sure the engines can read them, since most are "client side", this means their computers need to do all the work, instead of the web servers. This adds a lot of extra computing power for the Search Engines which could probably be better used elsewhere.

AVOID: Assuming All Sites Are Treated Equally -  Where is it written in stone that Google treats all sites equally? When attempting to reverse engineer some element of SEO, don't assume because something happens to one domain, that it will happen to all other, similar domains....That's just not the case. Often, a highly respected site will be "white listed", meaning they are given much more leniency based on a reputation of trust.

 

Text Vectors and N Grams - How Search Engines Detect Duplicate Word Patterns
Let's put this in simple language...As well as add some human "common sense" and circumstantial evidence to see if we can't figure out how Google may detect duplicate pages.

Text vectors are just "word patterns" or work phrases.

NGrams are: N = pick a number and "gram" meaning word...Or, how many total words in a text string have to be in common to be established as a match. For example, an N-gram of 5 would look at strings of 5 identical words to establish a match.

First, look at this document:

http://www.google.com/newsletter/librarian/librarian_2005_12/article1.html

Notice that it is on the Google.com domain...

It explains text vectors, which is just a fancy word for "patterns", and how Google creates vectors.

Next, we know that Copyscape is a partner with Google, as Copyscape is the same company that sends out Google's email alerts...Therefore, lets assume they have some influence and input and communicate with each other.

So how does Copyscape detect doops? It seems that text strings of 4-6 consecutive words start to trigger their dupe filter.

Therefore, we may assume that text strings of 4-6 words are "close" to a fingerprint.

However, this is probably too few words to be a full indication of a dupe page. For example, the string "cheap health insurance quotes" is a four word text string that is surely repeated on numerous NON DUPE websites, and shouldn't be enough to trip a dupe check, for obvious reasons.

But, how does this 4 word text string placed on a web page IN RELATION to other 4-6 word text strings? Or even a single word somewhere else in the document. And this is the text vector, AKA FINGERPRINT.

Is the phrase "cheap health insurance quotes" exactly 10 words ( a vector) before or after the phrase "office visits" and exactly 35 words before or after the phrase "no check up needed" on more than one document? If so, we're beginning to see some real fingerprints and possibilities for dupes.

Not only do you have to be concerned with 4-6 word text strings, you should also be concerned with ALL text vectors on a page.

Again, a text vector is merely counting the distance between two words (or word phrases) on a page, to create a number. Then, these numbers can be compared to other documents to check for patterns.

Conclusion: You need to have few REPEATED 4-6 word patterns, which seems to be the threshold of Copyscape (Google's partner), as well as remove as many common text vectors as possible, in order to beat the dupe filters.
 

Content, Making Pages and SEO
In reality, SEO/SEM all boils down to putting words on pages. Everybody has different standards and even different definitions.

What is a doorway page? To many, it's any page that isn't totally unique and hand written, designed solely to get traffic from the search engines, while other folks have different opinions.

So for clarity, let me give my definitions for the following terms:

Doorway - Any and all pages you create  hoping to get traffic from the SEs are "doorway" pages, by my definition. It doesn't matter if they offer great content or if they are totally spammy, they are doorway pages if you optimize them in the slightest. A doorway simply acts as a potential "entry point".

To me, a good "doorway" page not only brings traffic, it also offers a value to the surfer helping them find what they were looking for...Just like when searching in the phone book  for "foot doctor" and finding a suggestion to "Please see podiatrists", a doorway page can help the searcher by offering pages with different subjects where different words may be used to describe the same topics and subject matters.

Sitemaps and Hallway Pages - Sitemaps are pages with links to all your pages. Hallway pages is an "old school" SEO term that basically refers to sitemaps, with the purpose of hallway pages being to get your doorway pages indexed.

If you can picture a "hallway" leading to a "doorway", you can see how hallway pages will lead SE spiders to the doorway pages. I prefer the use of the term "hallway", because when we build them, we're not creating sitemaps in the traditional sense that the page is just a bunch of links to other pages. Instead, our hallway pages will be a little more discreet.

ConDoor Pages - This is a term I use that describes pages that are part content pages and part doorway pages. The key is to create pages that have real content on them for real people, but also act as doorway pages, optimized for powerful keywords to bring SE traffic. 

Duplicate (Dupe) Content - There are varied opinions about duplicate content. However, it is my opinion that the SEs do not want duplicate pages.

However, defining what is a "dupe" of another page isn't always black and white, with a lot of grey area. My guess is the engines would be more likely to use a percentage, such as "43% duplicate" or "12% duplicate", and factor this into their algo in relevant ways.

For example, if everything else was equal, and two pages had a "duplicate factors", where Page A was considered 0% duplicate and Page B was 19% duplicate, Page A would have an advantage.

We also need to remember that usually two identical articles posted on separate websites probably aren't exact duplicate pages, as each likely has different navigation, advertising or other elements on the page that would be different.

Instead of debating over what's a dupe or not, or if the engines want them or not, or the most effective course of action is to take...We first need to know how the engines detect duplicate pages before we can make a plan.

Caution: Use condoor pages on your Tin sites, which you use for testing and riskier strategies. However, don't be afraid of risk, as you can potentially drive tons of traffic to your site.
 

What's Missing from this Big Page of SEO?
This page is meant to specialize with On Page/On Site SEO, and doesn't include other important aspects of SEO/SEM, such as:

There's many other resources that cover these topics well, so they weren't repeated here.

I've created a page that lists the best free SEO, Linking and Page Building Resources called
The Big Page of SEO, Links and Web Building Resources.

Be sure to check it out as it is a treasure trove of free software, tips, scripts, articles and tools, all created to help you build sites and promote them through SEO and linking.

 

Conclusion
After giving thought to the future of SEO, there's really only a few logical conclusions we can come to about the future.

As the "spy vs. spy" game played by SEOers and search engines intensifies, the engines will need to react to the "billion blogs a second" programs and rely more on factors similar to YOU RANK and PEOPLE RANK to determine which pages are the "best".

So, if you accept YOU RANK and PEOPLE RANK, at least as a possibility in some form, then let's look at the solution.

We need something:

  1. That writes free content for us. Writes real, interesting content that is totally unique and original. Because it is original, it won't be considered doop by the engines.

  2. Gets Users to bookmark the sites and return often, so that PEOPLE RANK plays to our favor.

  3. Creates a site that can get good reviews from REAL human reviewers.

  4. Regular and frequent updates.

The logical conclusion that fulfills all of the requirements are forums, social bookmarking networks, "myspace" type sites, automation, etc.

While there are already major players in each of these options, the next level of creating major niches for each is still available.

For example, create a social bookmark site for the major niches, like "pets", "sports", "cooking", "software" and just about any other niches you can think of, concentrating on community type solutions that focus on User Generated Content.

Then create a "myspace" for pet lovers, as well as a forum dedicated to pet lovers, a digg.com type site for "pet" articles, etc. WordPress blog software has a add-on that allows you to give away blogs, which can be required to be about a specific niche such as "pets".

This User Generated Content can then be used to create "doorway pages" for SEO, content for newsletters, RSS pheeds and more, driving traffic and resources back to the Gold sites, consisting of forums, myspace sites, topical blogs written by Users, digg type sites, etc.

Simply, the SEO of the future will be built around the social type sites we now see, where Users generate the content. User Generated Content is the most "pure", legit content there is.

Here's a prediction...The major search engines will start to develop their own content. Why should they let all that traffic leak to other sites, when they can develop relevant information of their own?
 

Solutions and How to Effectively Exploit Many Strategies on The Big Page of SEO
I've created a number of programs and scripts to help squeeze every bit of power out of SEO, automation, RSS, content generation and much more. We probably have a solution and an idea or two for how you can take your website marketing to the extreme.

Featured in the Big Page of Tuelz Tips:

 
Free Tuelz Download Here - Includes: FatSearch, Moniterz, Updatez and The Newz Tuel.

 
Other Helpful SEO Inspired Programs by Kurt Melvin:

 

Do you wonder what has been added, modified or improved in this report?

Want to discuss or have questions about The Big Page of SEO?

Have an opinion, want to suggest a resource or discover insights into SEO strategy?

Then please join us at our SEO and Marketing Forum.

 

Thank you for visiting
 
Tuelz.com Copyright 2006 and beyond by Kurt Melvin.