In this document I will looking at an overview of the ESS reports and the data they contain. I will be explaining how you can interpret the results so you can do get your web pages in-line with the top 10 results. A full ESS report will give you written advice in each module but I will try and explain a little more about how that data is and advice is calculated so that you can have a better understanding of the overall report.
It’s important to understand how the Expert SEO System works. ESS takes your keyword and searches Google (or your selected search engine) for the top 10 results.
ESS then pulls the entire top 10 web pages and all their date into itself and runs a number of different analysis modules against that site data.
It does exactly same for the URL of the site you are targeting and entered into the project.
HOW ESS COMPARES THE TOP 10 DATA
With all the data gathered ESS can now compare your site against the top 10 result sites. Each module may create multiple measures and each measure is viewed in two different ways.
The range is the highest and lowest measure across the top 10 results. For instance looking at the number keywords in the body text may give you a range from 2 to 15. That basically means that the competitor site with the lowest number of keywords found in the body text had a total of 2 and the site with the highest keyword count had a total of 15.
So as the top 10 results have a range of 2-15 (for this particular measure) we can assume that Google find those numbers acceptable and even preferable for their top ranked web pages.
By comparing your own page result to the top 10 range you can tell if your web page need to increase or reduce it’s keywords to be in the same range as the top 10 competitors.
Of course you only have one page so your range will always be just one number.
The average is slightly different. If you consider that some sites may have a very wide range for some measures maybe from 0 to 1000s. That means that your site will probably be in range but that may not be truly reflective of the actual measures across the top 10 sites.
Imagine that the top 10 sites have measures like the following
1, 1000, 1100, 1200, 958, 832, 1026, 1181, 987, 1199
It’s rage would be 1-1200
Imagine now that your site has a measure of 2
Great news you are within rage. However if you look at the details of the range data it’s clear that most of the top 10 sites have a measure of around 1000. So your measure of 2 is actually not very good or representative of the real facts.
So we use also look at an average of the measures.
In this case the average of the top 10 measures is approximately 950
Now when you compare your site measure of 2 to the average measure of 950 you clearly see that you need to increase your measure
ASSESSING RANGE AND AVERAGE
So you can think of the range as something your site must be within. If it is outside the range it is clearly not what Google wants in a top 10 result for that keyword. So if your site is not in range it must be bought into range.
You can think of the average as your preferred target. It is not crucial to have an exact average but it should come near to that number. If it is considerably higher or lower than the average then you should get your site to increase or decrease it’s measure as required to bring it nearer the average.
What “considerably” higher or lower is will depend on the measures themselves. “Keywords In Title” for instance are likely to be very low numbers. Maybe even only 0 or 1 (it is or it isn’t). So even 2 or 3 may be well above average.
In those cases the range will be very small and the range measure will ensure you get your site correct.
On the other hand “Juice Passing Links” may have some very big numbers (even multiple thousands). If the average is 1000 for instance then a measure on your site of 800 to 1200 may still be “fairly average”.
So you can see that the size of actual measures may define what near average actually is.
One other thing to do in these cases is to look more closely at the individual top 10 data. An average number may be the result of an uneven spread (like the example shown earlier). The actual “best” number may be higher or lower but that will be more obvious when looking at the individual data.