First Sentence of Body Text

[moredial]

Tweet

Does that mean you agree or disagree with my point about tbp's "test" not proving anything?

Tests are very subjective. If you make a test and get A as a result, you repeat the test. If you now get B as a result you have to change the test slightly to see whether A or B is the more common result and eliminate the factors which caused the incorrect result.

Take that back about guru-e-books

I said, "It sounds like you have been reading too many gurus" not that you have been reading too many gurus.

Your arguments where like those I read in the sales pages where a bit of software will automate everything will save you from having to learn for yourself what does and what does not affect the SERPs.

There is no 'right way' in SEO. There is only a 'better-than-the-competition way'. If your content and your competition are all using prose styles which spread the phrases around like natural speech then that is what is 'normal' content and may do very well for a phrase which only appears once in the page towards the bottom. If you are using google, your page can do very well for a phrase which does not even appear anywhere in the content - not one word of it. And, such a page can rank higher than a page where the key phrase has a density over 10%.

If you make an experiment of adding your key phrase as the first indexable content after the body tag then you can let us know whether or not your rankings improved. We can't make the experiment for you because our content is not the same as yours.

All I can say is that over the last year I have been assisting one web master improve his rankings in a genre where SEO fads (rather than basic SEO) are more common than most as they try to outdo each other. The site was already doing well but he wanted to do better. The only change he has made to his previous SEO methods is to move his prime phrases first after the body tag and use correct html mark-up. He is so happy with the results that he has built another 5 or 6 sites, each one for a slight variation on the theme of the first site and they have all hit and stayed in the top 10 for some very competitive terms. For his first site he then did a complete layout redesign to use a search engine friendly layout used on the new sites - it looks the same to human visitors but contains about 10% of the total content. Better ranking, faster load time, more visitors, more page views, lower bandwidth. For him, the experiment worked - on more than one site.

He wanted to follow the SEO fads as they appeared to be working for the other sites. Fortunately he had faith in me and tested my suggestions on a few pages and saw that with lower densities his pages ranked higher than the latest fad - at that time, if I recall, it was adding tags to the page.

When this success is seen over and over again across different subjects the cause and effect becomes normal practice rather than just an experiment for those who use this one method to improve rankings. It is by no means the only method, it is just that it is the easiest and most predicable method.

[tbp]

As for my tests. I`m happy with the way my tests work, and I`ve seen consistent results. I`ve also shared the knowledge that i`ve gained to help others, and shared results that have worked for me.

If you don't agree with my tests, thats fine, its entirely down to you. Not going to argue about it anymore! I`m not forcing anyone to accept what I say as gospel, it's down for people to try things for themselves, but hopefully my own experiences will give a good starting point.

[Brendon]

Got any proof or is this more of the same old speculation presented as fact? I'm open to changing my mind if I see some of your evidence

Since you persist in being willfully dense, I will attempt to do some of your thinking for you.

Here's some evidence. It's a search related US patent from 1994. Let me quote you a choice paragraph

The sixth level in our default setting indicates that, all else being equal, give a record higher weight if it has the search words up front, that is, at the beginning of the document or record.

Is that clear enough for you? I'll freely concede that there is no objective requirement for words at the start of a doc to automatically be given more weight, but several decades of research and experience show that in almost all cases, humans tend to communicate something about the overall content of a document at the beginning of that document, where the purpose of the document is informational. Therefore, words appearing at the start of a diocument are more likely to be relevant to the overall document, and should be given more weight.

Simply giving a moments thought to the matter will explain why this should be so. Good communication generally requires that an overview of the content of a document is given first, so as to help the reader determine whether the document is likely to be helpful to them, and to provice a framework for interpreting the document.

This is NOT "the same old speculation", but published, verifiable research. There are plenty of other examples out there, but I found this one first. It IS (I promise you) an established part of Information Retrieval theory that in general words appearing close to the start of a doc are more relevant. Please don't further display your ignorance by attempting rebuttal by quoting a single counter-example.

IR theory is essentially a branch of statistics, concerned with using mainly statistical methods to analyse a large corpus of objects, and return the best apparent matches to the input query. As with any statistical discipline, it is always possible to find isolated examples that fall outside of expected limits. This has no bearing on the validity of the overall theory.

Similarly, works of fiction (which you brought up)

Newspaper bulletin style writing places important information near the top but creative, fictional writing doesn't. In giving more importance to phrases that appear near the top, the latter writing style would suffer.

contain very little information, from an IR point of view. They are not really suitable subjects for treatment by IR techniques because of their very different structural characterisitics. It's not really a weakness of IR, or search engines, it's just that those are not suitable tools for dealing with fictional writing, in the same way that a thermometer is not a suitable tool to measure distance. It doesn't make it a poor tool, just unsuited to certain tasks.

That doesn't make fictional writing any less valid as a writing style either, merely a recognition that the purpose of fiction is NOT to convey information. So, no, SE programmers don't care whether fictional excerpts are well served by SEs, since they have near zero value as resources, within the confines of the job they have to do

I was simply giving an example of a situation where giving more importance to content at the top of a document would break down.

Fair enough, but it's an invalid example

As for your ridiculous assertions about SEs having "many, many variables that can change in the background. We don't even know all the factors that search engines weigh up, so we certainly can't keep them constant. They are beyond our awareness and control."

Please. Take some time out of your busy life to

1) Do some basic background reading. Many of the important papers relating to SE technology have been publically available for some time. I thoroughly recommend Bill Slawskis' blog SEO by the sea as a good starting point. He has done detailed analyses of just about every significant patent app by an SE ever published, and links to many other pertinent sites

2) Do some research into research techniques. Not only is determining the variables used by SEs possible, it is also possible to produce a good working model of their algo by working with a sufficiently large dataset, and applying data analysis techniques. There is NOTHING magical about SEs.

In fact it's perfectly possible to make a first order prediction about SE ranking algos from first principles, simply by having a working knowledge of the WWW and human nature