Archive for the 'SEO Research' Category

I just read the Search Engine Journal post about the new Google PageRank Patent. Loren also references Bill Slawski’s blog where he gives us the facts and figures on the patent.

The name of this patent is “Method for node ranking in a linked database” (official patent link).

Abstract
A method assigns importance ranks to nodes in a linked database, such as any database of documents containing citations, the world wide web or any other hypermedia database. The rank assigned to a document is calculated from the ranks of documents citing it. In addition, the rank of a document is calculated from a constant representing the probability that a browser through the database will randomly jump to the document. The method is particularly useful in enhancing the performance of search engine results for hypermedia databases, such as the world wide web, whose documents have a large variation in quality.

As I read this patent I kept having to reread each sentence and translate it in my head since they write these so that most people will have no idea what they’re saying (except the lawyers “interpretation”)….so I thought I’d post my translation of just the Abstract of this document.

Everything indented below is Google’s words. Below Google’s words, I put my personal translation…I could be off…what do you think?

A method assigns importance ranks to nodes in a linked database,

Google’s algorithm tries to numerically value links,

such as any database of documents containing citations,

and pages/sites that link out,

the world wide web or any other hypermedia database.

to other web pages/sites.

The rank assigned to a document is calculated from the ranks of documents citing it.

We add up all the “scores” of those who link to you from other pages and give you a score based not on sheer number of pages that link to you, but rather how well “networked” you are within a community.

In addition the rank of a document is calculated from a constant representing the probability that a browser through the database will randomly jump to the document.

You get a higher score if you have multiple backlinks from related hubs/pages that are common for your industry.

The method is particularly useful in enhancing the performance of search engine results for hypermedia databases, such as the world wide web, whose documents have a large variation in quality.

This could help to give more value to sites that have backlinks from good related and/or trusted places within a common backlink neighborhood. 

———

What do you think? What is your interpretation?

———

Technorati Tags: google, google patent, link building, seo, sem, search engine optimization, internet marketing, search engine marketing

Popularity: 1% [?]

Jim Westergren published a very good link building guide today. (Well worth the read!)

I love most of Jim’s suggestions (he’s a very smart guy)…..but  I wonder about things like Link Vault and the Digitalpoint Co-op….I haven’t tried them…I often hear how wonderful they are (esp the Digitalpoint co-op)…..and sometimes feel like I’m missing out on something….but fear of google one day being able to map these out or finding footprints keeps me from going that route.

What’s your take on DP’s Co-op??

Popularity: 2% [?]

Natural Link Growth, The Famous Google Patent, Back to the Future thoughts.

In March of 2005 Google applied for a patent called “Information Retrieval Based on Historical Data“. If you haven’t read it, you really should…at the very least, read Rands “translation” of this patent.

Now I realize the some of you are saying “Just because they applied for a patent for this, doesn’t mean they are using all the points outlined in the patent application”…yea yea, I know…but some I believe they are using. Some of the points in that patent have to do with “Links” and “Time” I’ve outlined some of those items below (Keep in mind the “Historical” in the paper title “Information Retrieval Based on Historical Data”).

*Bold items were bolded by me, and were not bolded in the original patent application.

#22 ….wherein the one or more types of history data includes information relating to behavior of links over time; and wherein the generating a score includes: determining behavior of links associated with the document, and scoring the document based, at least in part, on the behavior of links associated with the document.

#24 ..wherein the appearance of one or more links relates to at least one of a date that a new link to the document appears, a rate at which the one or more links appear over time, and a number of the one or more links that appear during a time period…..

#25 …. wherein the determining behavior of links associated with the document includes monitoring at least one of time-varying behavior of links associated with the document, how many links associated with the document appear or disappear during a time period, and whether there is a trend toward appearance of new links associated with the document versus disappearance of existing links associated with the document.

#26 …wherein the one or more types of history data includes information relating to freshness of links; and wherein the generating a score includes: determining freshness of links associated with the document, assigning weights to the links based on the determined freshness, and scoring the document based, at least in part, on the weights assigned to the links associated with the document.

#27 ….wherein the freshness of a link associated with the document is based on at least one of a date of appearance of the link, a date of a change to the link, a date of appearance of anchor text associated with the link, a date of a change to anchor text associated with the link, a date of appearance of a linking document containing the link, and a date of a change to a linking document containing the link.

#29 ….wherein the scoring the document includes: determining an age of each link pointing to the document, determining an age distribution associated with the links based on the ages of the links, and scoring the document based, at least in part, on the age distribution associated with the links.

Below I do a bit of rambling….it’s part based on what I think the engines are trying to do….it is just some of my ideas I get from the patent.

When someone links to you, Google Dates it….and they also record the link text used to link to your site.  All this data is saved, and Google is trying to analyze it to determine ranking factors based in part on: The date someone linked to you with the first time. The date of the document that now links to you (I’d assume getting a link on a page that’s been up for years is much better than getting a link from a new page (90% of “links pages” I’d guess are less than 2 years old – another nail in the link page coffin ),

You can’t change your past. What has happened has happened.  I believe the engines are looking both at You’re past history, as well as where you’re at now. 

This original post was started by having someone ask a very hard question…the ole “How long will it take to rank in the top 10 for “XXXXX XXXXXX”.” He had a semi older site….had some backlink history…but few people were linking to him with the phrase he wanted…..problem is…is that lots of the people in the top 10 have had people linking to their sites with those phrases for years….how long will it take to rank for this guy now? All I can say is “probably longer than you might think” since they don’t have a record of any backlinks to their site with those phrases historically, and others do….and there’s nothing you can do about your past now.

With this person’s site I’d almost want to really study the link text of those who have been linking to them in the past and see if some of those are worthwhile phrases to target today (To find out use our Neat-o tool (often is down to overuse…but keep trying…there’s other programs also that can show this as well).

On a personal note, We Build Pages will probably never be in the top few again for the phrase “internet marketing” because I’ve got some shady history that’s been noted….oh well….we learn…and do better now.

Them big ole PR9′s and 8′s I bought back in 2002 that got the PR block in 2003…well I can’t change that…it’s my history…..them crappy 90% recriprial links I had in 2002….yea that’s my history….That hidden text in 2001….we’ll I don’t think google knows about that …them doorway pages back in 1999….we’ll they do know that those pages are long gone.  I’m sure there’s other skeletons in my closet as well…don’t we all have them?

Think about your link neighborhoods and link deals your dealing with today for the backlink history you write today might determine your rankings you hold tomorrow.

—————–

Technorati Tags: link building, google, links, link popularity, internet marketing, search engine optimization, seo, search engine marketing, sem, boykin, google patent, google, patents, patent

Popularity: 5% [?]

Funny….I was just reading about a conversation I had with Barry Schwartz 2 years ago. Back then I asked Barry not to disclose my name if he blogged about it for fear of Google coming after me for what I was doing……it’s been just over 2 years…and we’ve moved into what I hope are "better directions" than what I was discussing back then….but it was basically the beginning what what soon became known as "3 way linking" or "Triangular linking".

Popularity: 8% [?]

I was reading Loren’s article today, which referenced a forum thread on SE RoundTable.

In that thread Benjamin Pfeiffer was quoted as saying:

The best approach I think is rather instead of a site wide link a webmaster should approach a website by targeting individual pages and taking a slower more covert approach to obtaining links. I don’t mean hide links, but do your linking in a way that appears more natural. Place links here and there. This would thus make it more difficult to determine patterns in your approach.

This got me to thinking about "linking in a way that appears more natural", and blended with what Aaron was saying today…

Seems like Yahoo!’s related phrases / seo also section is based at least partially upon co-occurance of words in close proximity of one another across various web pages.

I’d bet my favorite cat, that Google looks at link text "co-occurance of words in close proximity of one another across various web pages"

…ok…let’s think about this for a second.   If you have 100 baclinks where each of those looks like this:

Jim Boykin’s Internet Marketing Blog
This is Jim’s soapbox where he rambles on about SEO and Internet
Marketing related stuff. Jim is also the CEO of We Build Pages.

Where the link text was always the same…and the next 4 words after the link text were always "This is Jim’s soapbox…" do you believe Google thinks this is a real "vote"? Do you think this looks natural?

Do you ever fill out one Form for submitting to 50 directories (same link text, same description?).

Do you do Link trading where your link and description after it is always the same?  (allow me to put another nail in the "link pages" coffin).

The words surrounding your link matter….this has been known for years…but rarely utulized by SEO’s.

What looks like a real vote…my listing example above done the same 100 times, or something like what is below?

….and Jim Boykin said "SEO Rules" bla bla….
….I was reading what Jim (an SEO) said about….
….quoting Jim (a self proclaimed SEO specialist)….
…that crazy Jim guy (he’s a SEO) link pages…
..an SEO named Jim said link pages suck….

Now, not one of those links above are for "SEO", but notice how the word SEO is near all the links.  I’ll ask you again, which is more natural?  Think I can rank for SEO this way? (I do).

Isn’t it kinda funny how SEO has evolved…in the past we’d find one "trick" and do it over and over again…the SEO of 2006-2007 has shifted to "how can I make this all look "natural" and not "tricky"?

———————

Technorati Tags: seo, sem, search engine optimization, search engine marketing, links, link building, internet marketing, link popularity

Popularity: 22% [?]

A deeper dive into Getting Ads on pages that have backlinks to them, and that value.

This post is a continuation of my last post which gave some ideas on pages to seek advertising on. One of the lines in there was:

…also aim for the “most trusted, powerful page” via searching Yahoo for [site:theirsite.com] and let Yahoo sort them by importance.

I want to just touch some more on this method of advertising…(call it building backlinks if you want, call it branding, call it advertising, call it “I want click throughs”, ….it’s all Internet Marketing to me.

….so where were we…oh yea, the ole “Yahoo Site: Command” (site:them.com) - What a wonderful thing indeed. (tip, don’t use the “www”).

Yesterday I was going over a prospective method of advertising on someone’s site, when I thought I’d check out the ole “Yahoo Site: Command” for their site.  There at #2 was the site’s Glossary….how good can a Glossary be? Let’s find out! So I went back to Yahoo to check the exact backlinks to that specific page. (in Yahoo use link:http://www.thesite.com/thepage.html)…yea, it was a Damn Good Glossary…it had had 740 backlinks to that page, including 44 .edu’s!  Yea Baby, that’s what I like!

I’m going to ramble on about some theories (or call it paranoia, or call it tactics) I have….

Let’s talk “Reputation” and “Trust”, using my definitions below.

Page Reputation = the words used to link to a page, the words used to link to the pages that link to them.

Page Trust = The backlinks to a Page and Where the Votes Originate (those trust values), and Trust Value of the pages that link to them.

So if you’re getting an advertisement on a site from a page that only has internal backlinks to it, then:
Page Reputation = The words used to link to that page internally.
Page Trust = Semi Low, since no one outside votes for this page.

On the other hand, if you’re getting an ad on a page that has outside backlinks to it then:
Page Reputation = The variety of words links in to that page, mixed with the internal words linking to your page…but External probably count for wayyy more.
Page Trust = Now your pages have a Trust value passed on from the reputation of those pages which link to that page, – this probably counts wayyy more than what is passed internally for Trust.

Let me show you a picture which I’ve drawn several times for my link team which can help you to visualize this.

You’re the bottom circle (the target). The circles in the middle represent the site you’re seeking advertising on. The circles on the top are the sites that link to the site you’re seeking advertising from.

The page on the far right has no outside link linking to it. It’s only “votes” are what the internal sites passes. (lower reputation and trust).

The first subpage you’ll see has lots of votes to it (links). The Glossary page I found had 740 backlinks including 44 .edu’s….Having an ad on a page like that (in the middle of the glossary, for relevant products services, yada yada,) would mean that the 44 colleges are directly linking to a sub page – they are directly voting for that page…the trust and reputation are through the roof….getting an ad on a page like that gives the trust and reputation a straight line to your target.

Ya, know…if you combine these links, with these links, you’ve got one Hell of a good start on link building.

Side note: Wanna know another reason why “links pages” are dead? Because those pages don’t have any backlinks to them from external sites! If all your backlinks come from “lower trust” pages (pages that only have backlinks from within the site, no external backlink votes), then guess what…you aint got Jack. I’d rather get 1 backlinks from 1 real page that has some direct external backlinks to it vs getting 50 backlinks from pages that don’t have any direct outside backlinks to them.

What do you think?

——————————–

Technorati Tags: link building, links, link popularity, seo, sem, search engine optimization, search engine marketing, internet marketing, google, yahoo

Popularity: 35% [?]

We don’t do Pay Per Click (PPC).

Who do you recommend for Pay Per Click Services, and why?

Anyone offer affiliate programs for PPC Services, or "finders fees"?

————-

Technorati Tags: ppc, pay per click

Popularity: 32% [?]