Like most webmasters I know, I used to think the way to learn analytics was to just dive in and do it. That simple method works amazingly well in most other areas of webmastery. Moreover, since the statistics one generates using a tool like Google Analytics are so impressive, it’s got to just be a matter of spending a little time with them, right?

Wrong. That way madness lies — or at least eyes-crossed by a sea of numbers. The usual pattern is you get your analytics tool to work, you’re enchanted by what you see, you poke around in the numbers, you show your supervisor, she is impressed too, and then what? You return a month later, and the numbers are up. That’s great, but you have no idea why and you’re smart enough to know that if you don’t know why, it won’t be too long before they go the other direction. One month later, the numbers are plateauing, and you still don’t know why. The next month you don’t go back. You’ve got better things to do with your time.

It seems to me (in retrospect) that it’s like trying to teach yourself finance by looking at ledger sheets with no one to explain them to you. Chances are you will see some things, but not enough to make it worthwhile. It will be an exercise in frustration.

And then the day comes when your boss says: “What about those statistics? Do you have something I can share with the Board?”

You could continue to give her the big picture numbers and be done with it. But I would suggest there’s a better way that will be a great investment of your time not only for your company, but also for your career. And it doesn’t even take that long.

I took one day each week in April to learn analytics. Specifically, I took the University of British Columbia’s online Introduction to Web Analytics course. It’s the first in a series of four, leading to the UBC Award of Achievement in Web Analytics, a certification recognized by the Web Analytics Association.

The result? To my surprise, with just this one course, I am (for now) Vanderbilt’s foremost expert at web analytics. That won’t last long. Several others are interested, and as soon as one of them takes the second course, my moment of analytics glory will end. But more important than that, I’m already finding easy ways to apply analytics. Best of all, as promised in the course, these simple things are already making management happier. And as you might imagine happy management makes a very happy web diva.

Of course, there are other ways to learn analytics. There’s an excellent book by Avinash Kaushik: Web Analytics: An Hour a Day. And there are plenty of tutorials on line. Just search YouTube. However, I’d caution you not focus too much on any one analytics application to start with — not even Google Analytics.

The first step is to learn the fundamental principles and practices. After that, the tools will make more sense, so you can use them more effectively. If you possibly can, take the UBC course. It’s first-rate.

To my fellow web professionals: I can hear you groaning. “I’ve already had to learn so many technologies,” you say. “Why one more? Give me a good reason for yet another responsibility.”

There are two compelling reasons to carve out some of your valuable time and learn analytics.

1. Using analytics moves website improvement past opinions into hard data. No longer will site design and architecture be based on subjective viewpoints — not even on HiPPOs — the Highest Paid Person’s Opinions.
2. It bridges the communication gap between senior management and web professionals. It speaks to what’s happening on your website and what can happen in terms that are much more familiar to management, e.g. data to back up the ROI (return on investment).

A favorite scene from a favorite movie:

Daniel: You know, Sammy, I’m sure she’s unique and extraordinary, but… the general wisdom is that, in the end, there isn’t just one person for each of us.
Sam: There was for Kate and Leo. There was for you. There is for me. [holds up one finger] She’s “the one”.
Daniel: Fair enough.
-Love Actually

I have the devil of a time picking what to focus on about the web. The problem is I love it all: PHP, databases, XML, CSS, Flash, convergence, standards, and so on. But for the past six weeks I’ve been focused on one that shows strong signs of trumping the lot: analytics.

I can’t begin to share all that I’ve learned in one or two posts. Instead, I’m going to take it in bite-size chunks. And today’s topic is…

What the Heck Is Analytics?

Once upon a time (measured in web eons, meaning back in the 1990s) analytics meant geek aracana like log file statistics, hits, page views. In other words, it was all about the raw numbers.

Then it evolved to mean the application that gathered the numbers — typically Google Analytics.

Now, however, as it comes into its own, it means much more. In essence, analytics is the process of matching core goals with objective numbers.

For those wanting a more thorough definition, it’s the practice of collecting, analyzing and reporting meaningful numbers about website usage. The data must be analyzed in an objective way through testing and measurement by those responsible for the site. Then the resulting information is used to optimize the site to more effectively accomplish the organization’s goals.

In other words, it’s powerful web juju.

After weighing the pros and cons of a number of excellent web conferences, in the end I chose the University of British Columbia’s Introduction to Web Analytics online course.

The deciding factor wasn’t the cost, the distance or even the speakers. It was the need of the Cancer Center. While I can’t say we have an expert at either search engine optimization or usability on the staff, nonetheless those are better covered then analytics.

And analytics really matter — in a way that reminds me of budgets. They are both about numbers, and numbers can be used to help management understand better what’s needed and why.

Needless to say, once I learn more, I will be posting more on this subject. For now, all I can say is get Google Analytics. It’s by far the best website number-cruncher I’ve used, and I’ve tried quite a few.