Two recent independent developments promise to shakeup the way traffic is measured on the Internet.  First Google announced plans to soon allow users to opt-out of being tracked through Google Analytics.  Secondly, the increased deployment of Flash cookies over HTML cookies has given analysts and business people a bit of encouragement, as Flash cookies are more sophisticated and harder to detect than their HTML counterpart, stoking the fears of privacy advocates wary of the amount of information collected on Google and search engine servers.

Anxiety Over the Future of Analytics

Google’s announcement poses some serious questions as to how traffic will be measured on the Internet.  Precipitated by growing concerns over privacy on the web, Google appears to be attempting to set itself in front of the issue and above the fray.  As Google Analytics is already an imperfect system, statistical purists find themselves asking whether data can be trusted once users can prevent their information from being tracked.  Once a significant number of Internet users opt-out, Google Analytics information may be rendered less useful.  Businesses will not doubt turn to other analytic services; however, the question remains whether other analytics tools will follow Google’s lead and offer users the same option.

On the flip side of the debate, analysts look to the promise of Flash cookies.  As mentioned above, Flash cookies are more sophisticated and pose hurdles for a user to detect and delete.  Furthermore, Flash cookies, one of the Internet’s better kept secrets, can re-spawn after deletion.  The additional hurdles users would need to overcome to free themselves from Flash tracking than simply opting out of Google Analytics could ensure that analytic information would retain its accuracy and integrity.

Not So Fast, Flash Cookies

Unfortunately for businesses and analysts, the public isn’t quite as ignorant to Flash cookies as they would like.  A recent article in OnlineMediaDaily claims that 7% of Flash cookies are deleted, which may not seem like a significant proportion; however, this number has doubled in the past 10 months.

When understanding the basics behind SEO there is usually a general theory that increased rankings will translate into increased revenue.  Is this always the case?  Maybe, maybe not.  We must first understand how we achieve revenue-generating conversions. 

The problem with many SEO campaigns is that only rankings and traffic are the core focus.  And frankly that is all many SEO companies report on.  Let’s take a look at Eric Peterson’s Web Analytics hierarchy of needs model to better understand the full spectrum of what is really important to complete the life cycle of effective online marketing solutions.

In this graph you can see that most campaigns focus on data such as ranking movements for keywords and the resulting boosts in traffic.  But what is this data telling us and how can we better understand what insights the data is providing?

Data can be exported into excel models or used in various analytics tools that will help you understand what is going on.  This is where you can really dig into each keyword (both broad and long tail) to comppare conversion rates and how that effects revenue.  That is why we need to really interpret the data and look for those key insights.

For example,  let’s say your website sells shoes.  You have two keywords ranking on page one of Google:  “men’s running shoes” and “men’s sport shoes”.  Consider what it would mean of “men’s running shoes” was generating 50 conversion per month in position number 3, and “men’s sport shoes” was generating the same number of conversion in position number 7.  That could possibly tell you that if you focus on increasing the ranking for “men’s sport shoes” then you could potentially drastically increase those conversion for that term. 

The idea is to have a system that is sophisticated enough to follow the path from rankings and traffic all the way to conversions and true ROI.  Once you have this system in place you can dial down on each keyword (and long tail variations) and assess conversions/revenue per keyword.  There are many different tools that can be used each step of they way, but it really takes a lot of manual testing and experimentation to get it right.