Sphinn Home » Web Analytics
A guest post by Aaron Wall on the Compete blog - one of these days I will have to go experimenting with the Compete data.
6 Comments     

Comments

from andrewsho 562 days ago #
Votes: 0 | Vote:
+ -

I have used the Compete data a fair amount.  It's pretty interesting stuff.  The biggest problem I have found is that it doesn't separate paid referrals from organic referrals so it's hard to get an accurate read on competitive SEO.


from emanuelh 561 days ago #
Votes: 0 | Vote:
+ -

AndyBeard writes: Compare those traffic stats against the algorithmic criteria needed to rank for those keywords and you have a good idea how much effort is needed to rank and how much reward each keyword brings.
I wish it would work. First of all you need, say, Google's algorithmic criteria but Google is very secretive so what we do know is at most 5% of what we would like to know. Unless you're a real novice and you still believe that everything can be found out in forums and blogs.

Secondly, if you collect data about your competitors at the top and examine it thoroughly, you'll find out that it is far from being arranged in a descending order of number of links, PR, age, etc. - the algorithmic criteria mentioned above. In fact a page with a very small number of links might be much closer to the top than a page with thosands of apparently good links. Same for PR, age and etc. These pages have therefore achieved their superior rankings despite and not thanks to their inferior qualities. We have no choice but conclude that they owe their higher relevance scores to thing we haven't measured, things that belong to the hidden information search engines also keep to themselves.

How much information is public and how much is hidden? I have argued in other posts that only about 5% of the information about the algorithm is public and that also only about 5% of the information about the relevance score of a particular page is public too. The unevitable conclusion is that SEO cannot be a technical profession. Read http://sphinn.com/story/18731 .
      

from AndyBeard 561 days ago #
Votes: 0 | Vote:
+ -

Note the Article wasn't mine, I just submitted it.

SEO is technical - there are aspects of SEO that I can't personally solve because I don't have the technical skill to create a time efficient solution, or I can't provide an easy solution for because my audience doesn't have enough technical skill.

from emanuelh 561 days ago #
Votes: 0 | Vote:
+ -

OK. So it's not you to blame. But you still have to demonstrate that SEO is a technical profession in comparison to my examples - electronics or automotive service - otherwise we refer to different definitions. And, to make things clear, SEO should be distinguished from website design and development (which is a technical profession).

An automotive service mechanic, for instance, is trained to perform periodic service to the car according to a definite set of rules and to repair the car if one of its components is faulty by testing their performance versus optimal performance standards. The goal and essence of SEO is achievement. It's much more like the race car driver's task versus that of the mechanic who services that car.

However, the problem of the technically-minded SEO is the absence of a definite set of rules like that which guides the mechanic's work. So it's not you're particular lack of technical skill that prevents you from pushing websites to # 1 for every search query you may wish but the collective lack of information on what's written in the textbook only Matt Cutts has apparently read.

from AndyBeard 561 days ago #
Votes: 0 | Vote:
+ -

Your typical automotive technician has a manual, tools, spare parts and a manufacturer to fall back on.

An SEO quite often has to make the tools, reverse engineer the instructions, and deal with a manufacturer who tries to be as ambiguous as possible other than saying "yes that is a car and it is broken" - combine that with an average client wanting a Fiat 126 transformed into a teleportation machine with a secondary time travel function and you think there isn't any technical side to SEO?

from emanuelh 561 days ago #
Votes: 0 | Vote:
+ -

It is true that the public (especially clients) look at the SEO worker as if he were a technician who can "fix" the site to be # 1 for travel or software. It is also true that this is the way SEO novices look at the future. The experienced SEO worker is usually more sceptic but he too has to find an answer to the question "am I an expert?", and the answer I provide is not very popular.
 
What is an expert after all? Someone who knows a lot. But in SEO you cannot know a lot because the search engines won't let you. So you either collect tons of irrelevant pieces of knowledge that are public and look technical to the uninitiated, or you become an expert in decision-making in conditions of minimal information. Which is something you cannot learn in courses, textbooks, forums or blogs. It is something to be learnt as an apprentice to a master. Well, you also must be a very critical apprentice.
 
And only after that you realize that the available tools for analyzing inbound links are wrong. They measure only what you are allowed to measure, for instance the sheer number of links but they can't help you find out which single link is worth something like seven hundred of the other links, or whether reciprocal links have a small negative value. For this you need a theory, a serviceable model of Google, and they won't let you have it.

And when you try to reverse-engineer the instructions, well,  suppose I want to find out whether 
Google's algorithm favors a certain property of the page, for instance the frequency of the letter Z in the text. So I devise a technique for increasing the value of that certain property. And now I observe the results:

My page was # 11 for a particular keyword and I implement that technique (by actually replacing every word I can with a synonym that includes the letter Z). And indeed, soon enough it jumps to # 9.

Hypothesis A: Google favors indeed this property and by increasing its value in the page you increase the relevance score.
Hypothesis B: The SEO workers of the pages formerly at # 9 and # 10 implemented a harmful technique (maybe they increased the frequency of the letter B). Their pages fell lower than mine because they were perhaps also much more hard-working than me!      


Log in to comment or register here.

Sphinn Sponsors

Be a Sphinn Sponsor - Click Here

Search Marketing Expo

Save the date for:
SMX Singapore - July 2-3, 2009
SMX São Paulo - August 4-5
SMX East - October 5-7, 2009
SMX Stockholm - 12-13 October, 2009
SMX Mexico - November 11, 2009

Search Marketing Now

Learn more about search marketing through free online webcasts and webinars from our sister site Search Marketing Now.

Upcoming Webcasts: