| 2:46 pm on May 1, 2001 (gmt 0)|
BH - We ain't ignoring you, honest!
I don't think there is a formal definition for keyword density. WPG may figure it one way and Brett's tool [searchengineworld.com] may do it another, for example.
It's an apples and oranges type thing. Decide what you're going to include when you calculate and do it the same way for each calculation. Being lazy and cursed with a poor memory, I include everything.
It becomes an issue if you're comparing your results with someone else's of course. Define the ground rules between you.
| 6:30 pm on May 1, 2001 (gmt 0)|
My apologies for not responding sooner. I had a server get hacked last week. Getting everything back in order definitely took away from my time here. :(
Regarding KWD, in the not so distant past, it was a fairly simple calculation of the entire page. I can remember coming up with good numbers by simply copy and pasting all visible text into Word, and then doing a word count and dividing the number by the number of times the keyword phrase appears. That used to produce fairly consistent numbers that you could use as a standard rule-of-thumb. However, there is now so many other variables, that I've found that it has become impossible to really establish any kind of solid KWD numbers.
The items that you should or should not include in any calculation will be different from engine to engine. That is why good KWD software will allow you to choose what page components you would like to analyze and then provide you with densities for each section.
I like to look at every component over several different pages. That can help you spot specific areas (link text, H1 tags, etc)that top pages have in common. The overal desities will often vary greatly, but you can usually spot a couple of areas that are similar.
| 6:40 pm on May 1, 2001 (gmt 0)|
So is it safe to say that WPG is just about the best, up to date, keyword analyzer software going?
| 5:12 pm on May 2, 2001 (gmt 0)|
I personaly prefer Keyword Density Analyzer by GRSoftware. It doesn't have any other features like WP, but it handles density great.
| 2:18 pm on May 3, 2001 (gmt 0)|
WRT Keyword density analyser is there any opinion as to which choice to use to analyse a page, sum occurences or whole phrase ?
The whole phrase does not seem to give the right numbers, does google rank pages on the whole phrase scenario, even when a searcher does not include the query in speech marks ?
My feeling is towards sum occurences.....what do you guys think ?
| 10:52 pm on May 3, 2001 (gmt 0)|
It really depends upon the engine and whether or not the default to exact (proximity) matches.
For an engine like Google that really doesn't seem to care whether or not the words in a search phrase appear on the page in close proximity to each other, the Sum setting works best.