A prominent theme of one of my favorite books, Zen and the Art of Motorcycle Maintenance, is the juxtaposition of two worldviews classified respectively as “Romantic” and “Classical”.
The Romantic viewpoint, tangibly represented by a shiny new BMW motorcycle, is one that does not obsess over details and has little desire to attain deep understanding of life’s motivations. Romantic practitioners favor polished systems that require little or no maintenance and prefer not to dirty their hands by digging inside those systems.
On the other hand, the Classical viewpoint, represented by an old Harley Davidson motorcycle, is one that is built upon attention to detail and analyzing every aspect of life. A Classical perspective is one that prides itself on full control and the ability to experiment and tinker independently to overcome life’s obstacles.
In the book, the Classical viewpoint is clearly favored. The author is constantly tinkering with his aging Harley Davidson and on several occasions his hands-on knowledge of the vehicle permits him to diagnose and repair it independently. In contrast, when there is a problem with the BMW motorcycle driven by the author’s acquaintance, professional repairs are required to fix the bike.
Let’s take this metaphor to the digital world. One of the great things about digital media is that tools exist to track user behavior. Tags, cookies, beacons and pixels all allow for a publisher or advertiser to see how people are interacting with their content or ads. Although the variety of available tracking tools is fairly standard, the application of those tools is currently a purely Classical system. In some ways it’s good that applying online tracking requires a lot of tinkering and hands-on knowledge. However, on the other hand, the practice as a whole is rife with inconsistencies because practically each instance of online tracking is a unique implementation. Even the seemingly insignificant difference of placing a spotlight/action tag at the top vs. the bottom of a webpage can lead to a 10%-30% discrepancy in a tracked metric.
Digital analytics and research, which relies on the data collected by online measurement, is (unfortunately) always influenced by the implementation of the often fickle online measurement tools.
When you’re riding down the highway on a motorcycle it’s good to know that you can tinker with your bike and repair it on the fly if necessary, however when it comes to the practice of digital analytics – which relies very heavily on the ability to isolate similar test and control groups – it’s far better to rely on a system that is slightly more structured and rigid.
In today’s world, due to the discrepancies in online tracking implementation, online research requires the comparison of campaigns that can be as disparate as a roaring pack of Harley Davidson’s. However, as technology progresses – and online tracking becomes more reliable and consistent – it will be as if we are all riding shiny new BMWs in perfect formation – which will allow for far more accurate online tracking and research.
I suppose the question is – who will create the BMW of the online tracking world?