Two of the worlds leading experts, Avinash Kaushik and Eric Peterson have been having a long term discussion on whether analytics is easy or hard. Am I going to answer the question once and for all. It is both.
Now you might be wondering how I can claim to know the answer when compared to these two I am a total noob in the industry. The answer is because I AM a total noob. I can see the parts that are easy and the parts that are much more difficult. I suspect that for Avinash this is all easy because he has been at it for so long and for Peterson it is diffucult because he doesn’t even see the easy stuff anymore, its just background noise. At my level of experience it is all…there.
For example some of the basic concepts of analytics are simple to understand.
- baselining – How are you going to know you’re improving if you don’t know where you are starting from?
- A/B split testing – Take two samples of an element and serve one half of the time, and the other the rest of the time. See which one does better.
- ROI – return on investment – For every dollar you spend, how much do you make back?
- Importance of corporate buy in or executive sponsorship
- data driven business decisions
The list goes on…
And web analytics is easy to integrate into your company. I generate a simple report with Google analytics to report on a variety of metrics. That report can arrive every Monday morning on the desks of every executive in the corporation. But here’s the catch…
Doing something with web analytics is hard.
Web Analytics is hard to implement properly not only from a website standpoint but implement into corporate culture.
Anyone who has tried to implement and integrate web analytics code into a site can attest to how difficult it is. Not only because most of the time the code is fairly complex, but also because it is hard to make sure it is accurate. It is hard to find a company that understands the value of web analytics, it is hard to get executive buy-in. It is hard to tell a client to stop doing something that wastes money when someone else’s career hinges on that project. It is hard to do run a linear regression test in SPSS. The list goes on…
While I doubt Avinash and Peterson will ever see eye to eye on the issue, I appreciate the fact that there people out there willing to share their expertise.