Summary of SEO - What and What Not (My beliefs on SEO)

[adesbarats]

Second there isn't much controversy AT ALL that Meta tags are of VERY little value. Thats widely known and been confirmed on mutliple occassions by Google so that part of your post was strange.

Mike,

I would like to just take the opportunity to say that meta tags are probably VERY important from the standpoint of achieving click through. I am noticing in many threads that people are very focused on tweaking their title tags and meta tags to get themselves on the first page but there seems very little discussion about how that impacts their click through rates. For example, let’s say you are trying to rank for the keywords “weight loss”, “diet plans”, “diet supplements”. From an SEO perspective you might do something like this:

Title: domainname.com – Weight Loss Plans, Diet Plans, Diet Supplements
META: Buy your diet plans, weight loss plans and diet supplements online with fast, friendly service and money back guarantee….


Now, the above title and meta get all the keywords you are targeting so from an SEO perspective you might feel you’ve done a great job. But have you? Maybe you get on the front page but we tend to forget that the title and meta are the first thing people will see when they search one of our keywords and hence they must play a role in enticing the prospect to click through to our web site. Sure, I know that on average 50% of organic traffic clicks on the top rank, 25% splits b/w #2 and #3 and the remaining 25% is evenly split b/w the next 7 spots or something like that. But that is just a very general rule of thumb that I am sure you could skew in your favor with a well thought out title and meta that is both SEO friendly AND marketing friendly. Say something like this:

Title: Want a Weight Loss Plan or Dieting Plan That Really Works? Diet Supplements that Burn Off the Fat? (notice I removed the domainname.com as it is a waste of space IMHO)
Meta: Tired of trying all those diet or weight loss plans that just don’t work? We guarantee you will get the weight off & keep it off or your money back!

Which one of the above is the more powerful ad? So, I would agree from an SEO standpoint, the meta likely has little value but from an advertising standpoint, it IS extremely important! As such, it should not be ignored…

[FredJones]

The Sum-of-Product in fact does appear in this paper (The Anatomy of a Search Engine), and in the same way that I describe it.

Let me show how and where to anyone reading this, so that Warriors not habituated to read Scientific papers also understand it in their right context.

Head over to section 4.5.1 which talks about the ranking system.

It says: "We take the dot product of the vector of count-weights with the vector of type-weights to compute an IR score for the document. Finally, the IR score is combined with PageRank to give a final rank to the document.".

The pagerank part is mentioned here (and anyone with the slightest idea on such computations would say it is a multiplicative factor rather than additive, though in principle the word "combined" that they used is vague). Now let me head over to the other factor mentioned here - the IR score.

Now, look back at the IR score. It contains (a) the on-page factors as we call them and (b) the anchor.

Notice this from section 4.5.1: "Google considers each hit to be one of several different types (title, anchor, URL, plain text large font, plain text small font, ...), each of which has its own type-weight. The type-weights make up a vector indexed by type. Google counts the number of hits of each type in the hit list. Then every count is converted into a count-weight."

The word "anchor" appears here, as a factor. And note that there is a different point in the space for every anchor ("each of which has its own..."). What does that translate to in plain English? It translates to, in my understanding, "each backlink with the right context, that is, the right anchor text, counts to rank the page for that term".

Now, to what extent does that "count"?

Go back to what I had started quoting: "The counts are converted into count-weights and we take the dot product of the count-weights and the type-prox-weights to compute an IR score."

In plain English, a dot product is a sum-of-product (the same thing that I had explained), when it comes to a vector.

and pretend that your calculations are in there

To clear up any confusion, I am not "fudging", "pretending" or "hiding". These are Brin and Page's (founders of Google and the original inventors of their search technology) calculations and not my original ones. I don't even see how could I hide behind their calculations. I just apply my understanding of this calculation and it works every time as I expect it to. That's all.

But expressing terms such as "dot product" etc in simpler terms such as sum-of-products in the way that I have done ought to help Warriors - just my belief.

[Mike Anthony]

Fred I don't know how to tell you any more simply than I already have that you are harping on the things to which there is no controversy. Maybe you should familiarize yourself with those who do SEO more. You act like you are revealing water for the first time to a sailor.

Anchor text weight, Keyword density, the placement of Title and H1 tags, Placement and LSI are NOTHING NEW and never required of many SEO to study computer science (although theres nothing wrong with that). These have been known by people who actually rank sites for years. Those are PRECISELY the things that 4.5.1 refers to AND THEY ARE KNOWN BY EVERY SEO ON THE PLANET. We talk about them every day in this forum and frankly we do it in a way far more approachable than you have laid out.

However you make two critical mistakes.

A) You take partial revelation as full revelation. The purpose of the paper is not tell you how to rank. Google is not spilling all the beans. They are talking in broad concepts not exhaustive terms and the main function of the paper is broad not specific to the algo. So no it does not endorse your claims of being ale to rank for any term number one by following that incomplete algo. It is VERY WELL KNOWN that Google has MANY factors it does not discuss. IN fact he paper hints at that being the case at the time and in the future (as far as the paper is concerned).

B) you extrapolate from what they do share into your opinion and confuse the two.

I have made it clear what I disagree with when you do this. I called it a smokescreen because I told you in my very post that I didn't disagree with everything and I asked you to specifically give some evidence for there being Contextual PR separate from plain PR from which you derive your questionable equation (which is NOT in the paper) in post number two. I am stillwaiting for a link to that from Google. Unfortunately so far you are hiding behind not having to provide evidence for anything that is questionable.

So at the moment there is no new light being shed by your reference to the paper. Theres not a single thing in it that every seo doesn't already know and it doesn't address what my issues with you making up additional algo facts

Worse it illustrates how unsubstantiated your claim is that you can rank number one for any term. If all there was to it was what was discussed in the paper then SEos all around the world have known and been using it for years which would make being one for every term impossible. Everyone can't be number one using the same model.

Therefore your claims remains absurd. You cannot rank for almost any term you chose number one. I've said this several times and rather than withdraw the claim or clarify it you persist in claiming it while claiming you need show no proof

Have you had the opportunity to read the "historical data" links? Notice all the factors that are not in the paper you refer to. At some points you complicate things too much and at others your assessment is to simple.

[FredJones]

Originally Posted by Mike Anthony

Have you had the opportunity to read the "historical data" links?

Yes. Thanks for adding that to the thread.

[pandorasbox]

You know it is very funny, I was having this same argument a bit ago. A website had a page rank 4 and like 5000 links but wasn't ranked well for anything. Everything must be yin-yang ballanced. The foundation for SEO is on-site and if not done properly you cannot do any type of link building correctly. I think that even sometimes on-page is more important in almost every aspect than of-page because to many people build a massive amount of links and get no place fast but if you build niche links and relevant links you can get to the same place much faster and pass your competition that is building thousands of links. Anyways, this was a great read. There is a big difference between doing SEO correctly and saying you are an SEO expert. Most people say they are but most are not.

[FredJones]

Lol, heartily agree to this one and see the funny side also. Too many people equate SEO with bum backlinking, and of course SEO is much more than that

Originally Posted by pandorasbox

There is a big difference between doing SEO correctly and saying you are an SEO expert. Most people say they are but most are not.

[Mike Anthony]

Originally Posted by adesbarats

Mike,

I would like to just take the opportunity to say that meta tags are probably VERY important from the standpoint of achieving click through.

Are you referring to met tags or meta descriptions. Meta descriptions are still important and Google will use them to display in the serps.

[NicholasX]

Anybody knows how to get high PR Backlink??