What does it take to be a Web Analyst?
When I first heard about web analytics, I felt that it would be the perfect merger of my business background and the always growing, ever-changing Internet world. While I didn’t have all of the technical terminology down (according to my web developer husband), and probably still don’t, I knew I had the statistical, reporting, detail-oriented, and strategic skills to at least get me started. Plus, I was looking for a career not a job, and with the growth state of web analytics as more and more companies looking for marketing budget accountability, it seemed the perfect fit.
So now, a year or so into my web analyst studies and in the midst of report development for a local marketing software firm, I thought I would simply outline a couple of things that I’ve found to be important in just this short time. This list is neither comprehensive nor an expert source, obviously, but hey, I still have time!
1. Requirements Gathering
As I was looking for tutorials on how to write a functional spec (the document that details every aspect of a site for the developer, to the point that they have NO questions), I found the phrase that the spec (as well as the analyst, in my opinion) involves the “marrying of business strategy, technical requirements, and usability,” with emphasis on customer/client usability above all else. While you may think first of the business needs, i.e. what you need to gain or provide with your site, those needs are meaningless if customer cannot navigate, read, or use the site effectively.
2. Presentation & Visualization
Sure, Google Analytics or SiteCatalyst may have the fancy, built-in, interactive charts, but when I was tasked with coming up with our own system of reports for use by marketers and analysts, I had to think about every chart and graph in terms of clearly presenting the data in the way that makes it almost obvious what action to take or interpretation to make. I will not be there myself to explain the data or make educated recommendations, so I have to shape the reports so they can be read anywhere from the most to the least experienced analysts. Let the experienced analysts have the raw data!
Obviously, charts are great, but can definitely be subject to “data overload”…just as companies who do not know how to use the immense amount of data they are provided on a daily basis, it is best to keep it simple. By focusing on data that drives ACTION versus every known fact, figure, or chart, users can stay focused on the goal at hand and selectively drilldown to areas that need further investigation.
3. “Selling” the Data
After gathering all the business and technical requirements, then deciding how best to present the data, one crucial aspect will likely be selling your “work” to SOMEONE–whether it be getting your project manager to sign off, getting a client to use it, or justifying an action you took based on the data. Unfortunately, while there may be “industry-standard” terms, there is rarely a standard definition of what the term means, or at least how the data behind it is collected and measured.
The least you should do is come up with a consistent definition and stick to it. Then, at least those reading your report can understand where you are coming from…even if they don’t agree with your definition, they can at least trust that you results are nothing if not consistent! If they want to argue the merits of your calculation later, at least you have a leg of reasoning to stand on.
Web Analytics Demystified Blogs (by such WA gurus as Eric T. Peterson and Judah Phillips among others)
...and of course you can follow me on Twitter where I post about my searches and often "re-tweet" other great resources!