After having developed (with the help of numerous blogs and books) a my web analytics strategy over six months ago, I am faced with a new challenge - the reality that a one-size-fits-all strategy doesn't always fit.  The "bodies" are not always in "alignment" in order for a goal-based strategy to operate.  And sometimes, pages/sites need intensive analysis before a goal-based strategy can be valuable.

Now I will contradict myself.  A one-size-fits-all strategy doesn't always fit unless it is a living, breathing organism that you are willing to restructure as you go along.  By adding a new "step" to my current strategy process, it will now fit all of my current challenges, thus making it a one-size-fits-all solution once again.

So what's this "new" step?  One-time, thorough, intensive analysis; or, for short, intensive analysis.  I have placed this optional step at the beginning and end of my current process:

Ok, great, so what is inside this new box?  I put together a preliminary draft of what I'm including in my intensive analysis.  Of course, once I start using it, I expect the list and it's order to change drastically as I'm sure I will realize some obvious territory that I neglected to include.  Here is my intensive analysis list with some scribbled notes on the content of each:

1. Visitor Survey Analysis - Numbers can be picked apart, but when the customers speak, they're clear.  I pattern my surveys after the 4Q survey (Thanks, Avinash Kaushik).  It's important to get your survey up right away.  You don't want all other areas of analysis to be over and then have to wait weeks for ample responses to be collected.  Try connecting visitor feedback to metrics for stronger recommendations.  Multiple recommendations may come out of this analysis.
   
2. Heuristic Analysis - This basically means to learn about the page/site from the creator and the visitors perspective.  One area I recommend looking into is page size compared to visitor browser size and desktop size.  Another idea is to gather a list of best practices and compare the site to them.  You may not come up with any recommendations with this analysis.  The main point is for you to wrap your head around the intended purpose and architecture of the site and what the visitor sees.  It's a learning experience to prepare you for the remainder of your analysis.  Most Importantly, many ideas will pop into your head of potential areas of analysis you should perform.  Jot these ideas down and see if they fit into any of the following analysis categories.
   
3. Create Visitor Profiles - I thought of this one while halfway through this list.  Try to understand the different types of visitors to your site and create a sample profile of each type.  Keep in mind that this is just another way of being in your visitors' shoes.  Eric Peterson's book, Web Analytics Demystified, has a short section on this topic that provides a nice introduction to the topic.
   
4. Conversion Analysis - Determine what conversion(s) this page/site is concerned with and analyze the performance.  Study the conversion path and make recommendations accordingly.  There is a plethora of information available about conversion analysis.  Refer to it as necessary and then some!
   
5. Revenue Analysis - Determine (or approximate) the revenue stream of this page/site.  Approximate?  Yep, some sites are strictly for the sake of providing content (as with many of the sites I support).  However, there is usually a way to connect them to revenue even if it isn't one of their goals.  As I always tell them, "You're gonna be asked about it sooner or later; and if not, then you'll just make yourself look even better."  It's a win-win!  Execs understand dollar signs (even though they may talk about volume all day long)  ;)
   
6. Incoming Traffic Distribution Analysis - Where does the traffic originate from?  Look for areas of improvement.  Save your digging around in search traffic for the next analysis.  Segment your various reports for typed/bookmarked, website, or search engine entries.  If you want to get really crazy segment various reports for specific websites or search engines.
   
7. Search Engine Analysis - Put on your SEO hat!  Every web analyst has a little SEO in him/her.  Do I even need to detail this?  Nah!
   
8. Content Usage Distribution Analysis - This is something I got from Avinash.  How much of each content type do you have versus which types of content are your visitors spending the most time?  A combined column distribution graph will tell a ton about what you're not providing enough of (or too much of).
   
9. Navigation Analysis - Where are visitors going on your site and how are they getting there?  Is your navigation confusing?  You could spend a week on this!
   
10. Competitive Analysis - This is probably the least plundered area of web analytics.  Just looking at what all the web analytics blogs talk about reveals how little this area has been studied.  We all know about the tools out there, we just need a little nudge to start learning how to apply them (myself included).  This is last because there's no sense in studying the competition if you don't understand yourself.  How does the competition do things?  Can you improve on an idea of theirs?  Be sure to provide an overall idea of how your website is doing compared to the competition.  Everyone has a competitive streak.  :)
    

Many of these items overlap.  You'll have to decide to dig in now or write it down and save it for later.  In nearly all of these steps, you should end up with at least one recommendation.  If you have multiple recommendations, order them by considering importance, ease and ROI.  In the end, provide a complete list of recommendations in the order you think they should be considered.  You should come up with nine or more recommendations.  This intensive analysis could be done in as little as a week (if you have nothing else to do or you aren't "intensive").  Realistically, I expect that these will take me two to four weeks.

I'd love to hear feedback about other areas I missed or if you've put together such a list yourself.