What am I suggesting?

Simply that you do not count people who were coming to your site anyway  even if they come from Google as part of your Google SEO success.

Why do I say this?

I am often told by people that 80% of there visitors come from google and that they would be out of business if Google dropped them. 

However when I investigate further or get them to investigate there own stats it usually turns out that Google only accounts for 60% or less. On one of my more successful sites (avg. 100,000 uniquest a month), Google is only truly accounting for 34% of unique visitors, despite 67% of traffic coming from Google.

Why do I say this?

Simply because its a unique branded domain, when people search for the name I know that they are looking for my site and not some product or service, this combined with the number of people who partially type in part of the domain name convinces me that these uniques were heading for my site anyway.

Prove it! 

Here's how ow my stats break down based on 100,000 uniques with 67% of traffic coming from Google.

 

*(All numbers were rounded) 

So when I am looking at my Return on Investement model for the above site I see:

• Brand Marketing  and non Google SEO as generating 66% of traffic
• Google SEO giving me 34%  of traffic

While this site is not what I see across all my others it does prove to be a useful example in explaining why I think that you must exclude certain traffic that google sends you from your Google SEO budget and include it in your Brand / General Marketing Budget or non Google SEO Budget.

Caution!

Now obviously I want to be no.1 for my branded terms and it will hurt if someone else is above me, but this is essentially google getting it wrong rather than driving traffic to me. Tthey were coming to me anyway but google could potentially send them somewhere else. This just requires basic defensive SEO to make sure that I am listed top for my brand, I do not believe that this traffic should be seen as a Return on my Google SEO but as a return on my branding and marketing as thats the reason they typed the term in, until Google is an awful lot better I will always have to have some cash set aside for defensive work I need to do to make sure they get it right.

Summary:

So in short, when looking at stats you need to discount keywords that are your owned brands or ones where people have just been lazy and typed your url (or portion of it) into google rather than the url box on your browser, its different for every site but the basic premise is try to discount people who were coming to your site anyway.   Obviously you should treat all sources in the same way, I hope you get the idea.