As a search marketer and analytics geek, I have one bit of advice to all other search marketers out there - make sure you get the chance to analyze your client’s traffic logs yourself. Otherwise your search campaigns may not get the credit they deserve.

Consider the most common scenario we’ve come across in our agency. Let’s say a client has been running multiple simultaneous marketing campaigns, including pay-per-click search, SEO, and offline channels.

Also, let’s assume you’re using trackable URLs for the paid search component, and Referrer data is included in the log files. The traffic driven by each tactic should be obvious in the client’s metrics reports, right?  

Wrong. 

• Repeat visitors may be driven by multiple online/offline ad exposures
• It takes extra effort to track latency - the lapse of time between the first visit and a subsequent visit that generates a conversion
• The dreaded "No Referrer" creates a fog because all these clicks are assumed to come from offline campaigns.
• Clicks coming from portals such as Earthlink or CNN often aren’t counted as search traffic when in fact these and many other content sites are in the Google and Yahoo/Overture networks.
• Some search traffic appears to come from unresolved IP addresses that your software can’t identify unless you manually assign those IP’s as search engines (I find this happens often with Google and MSN)
• Clients may find it easier to consider "Top Referring Sites," instead of "Top Referring URLs."  This can make it hard to distinguish between paid and natural search referrals.

So what’s the solution?  (see my next post Measuring Search ROI Is Harder Than It Seems - Part II)