One of Avinash‘s favorite expressions is “data puke’, which goes a long way to describing the vast majority of charts, tables and graphs that I’ve seen over my relatively short (but I like to think illustrious) career in web analytics.
One of Avinash‘s favorite expressions is “data puke’, which goes a long way to describing the vast majority of charts, tables and graphs that I’ve seen over my relatively short (but I like to think illustrious) career in web analytics. I’m no “PowerPoint Picasso”, but I believe that I can put together slides in a manner that imparts knowledge, disseminates data and tells a coherent story.
Evidence indicates that I and my colleagues are in the minority. Perhaps this accounts for why many big corporations seem so very keen on the concept of web analytics, yet so very reticent to actually do anything with their data. They’ve either paid big money for an analytics suite that comes with a consultant, or they’ve used a free solution like GA or 4Q and hired someone as a web analyst. You’d figure that they would take advantage of all the available data, considering the money they’ve spent gathering it.
Personally, I think this corporate inaction can be tracked back to three key things.
Collecting a database of all your visitors’ on-site activities generates a pile of records that is nigh on impossible to parse through, even with highly-paid analysts dissecting every keystroke. Even when you think you’ve reached a conclusion, a simple re-segmenting of the data can show you something different. Everyone thinks that all this data will highlight the Yellow Brick Road that the company should follow, but more often than not you simply end up standing at a 4 way crossroads, spinning in place and wondering which way to go.
Related to the above, all this data keeps everyone stuck in the same place, unable to move in on direction or another. You’ll get everyone in the company agreeing that something has to be done, but all the time will be spent trying to figure out what. Typically, you’ll end up with two camps, each with opposing conclusions that are backed up from data drawn from the same source. The net result is that decisions take forever, if they come at all.
Certain things in the corporate world are set in stone. Deployment schedules are one of those things. While they may have been put in place to keep the company running smoothly, they typically ensure that all the company does is play catch-up. When you receive feedback, you have a very short time to respond – yesterday’s news is old news, and if you are forced to wait 2 months for the next release date, an opportunity will pass you by.
Whether it’s too much data, indecision or corporate procedures, data obtained from various web analytics sources is not acted on quickly enough. Delayed action results in reduced returns and the perception from many corporate higher-ups that web analytics just isn’t worth it.
Deployment schedules need to be thrown out the window for smaller items – with releases coming weekly, if not daily. Quicker turnaround will offer greater rewards, and increased ROI isn’t something that needs to be explained to the HiPPOs upstairs.
This is a repost from Christopher Pam’s blog. The original can be located here.