Is Web Data a Science or an Art?
Baseball season is finally back and I’m pumped. I’m a HUGE baseball fan and one of the big debates going on in the game in the past few years centers around the use of statistics (namely Sabermetrics) to measure how good a player is. Traditionally, scouts would comb through players around the country—watch them play—and decide how high to draft him and what he was worth.
But now statistics are being commonly used to essentially replace the role of the scout. The scouts—long the bastions of locating talent—are understandable upset.
Instead of determining the value of a player by watching him play and making a relatively subjective decision based on a scout’s experience, instinct, and knowledge—more and more teams are crunching a bunch of numbers together and letting the math do the work for them. That’s why Moneyball caused such a stir when it came out: it was the first time baseball fans realized how big of a role the math guys were already playing in the decisions about which players were worth what.
The same concept applies when you’re talking about your website. Of course the numbers matter—otherwise this site wouldn’t exist! But relying too heavily on the numbers and the story they tell is dangerous (most companies have the opposite problem, unfortunately). If you ignore your experience, judgement, and subjectivity by deciding to submit yourself entirely to the numbers game, you are in for a world of pain. In order to get the most of our your site, you need to understand that, as boring as the term “web analytics” may sound, it’s a combination of art and science—not just about number crunching. Just like good baseball teams mix Sabermetrics with scouting to determine value, that’s how you should make decisions about your site: by combining what you see with what the numbers tell you.
It’s easy to fall in love with the numbers: they don’t lie (unless you are reading them wrong or you implemented them incorrectly) and they’re pretty easy to compute. They also look great in a chart, graph, or Powerpoint presentation. If a decision fails based on the numbers, no one is to blame. You can always say “the numbers failed.” But if you fall in love with the numbers it will be an expensive, time-consuming journey for you and your site. It takes time and money to set up systems that will measure exactly what you want to see. Even installing free software like Google Analytics, as easy as it is, takes time and some knowledge.
Say you run a site that sells T-shirts to college students with funny lines on them. And then your boss comes in one day and says, “You know what we should do? We should shoot some videos of people folding their tshirts in their closets or putting them up on hangers. You know, to like show how people store their tshirts? It could go viral! What do you think?”
There are two things you could do here (although you should probably look for another job first):
- You could set up some kind of system on your site to gauge user interest on this type of thing. Track it, crunch it, measure it, and reach a conclusion.
- You could just tell your boss that the idea is stupid.
If you’re in love with the numbers, you might want them to make the decision for you so you don’t have to confront your boss and tell him what a moronic idea this is. That means you’re going to spend the time, effort, and money on creating a system to gather those numbers. Then you’ll spend the time to analyze the numbers. Then you’ll waste everyone’s time by calling a meeting to discuss the numbers.
Not very efficient. Use your subjectivity instead. Just like a scout could easily tell you that Ryan Braun had no future as a third baseman without crunching a bunch of numbers, so can you. You have to trust your judgement so you don’t over rely on numbers and so you can get things done quickly.