by Dorota Umeno on January 16, 2011
I was reading the latest blog post by one of the undisputed authorities in web analytics, Avinash Kaushik. He wrote a post titled: I Wish I’d Known That. [Digital Analytics Edition.] In it Avinash says:
Everyone wants the perfect tool that bounces rates in real time while computing multi-channel attributable impact of an email sent to grandma via Facebook based on competitive intelligence gleaned from TV watching behavior of customers with lifetime value greater than $358 and FICO scores of 700 or higher.
Get over it.
I chuckled in agreement. We (digital strategists) have certainly become increasingly greedy for “more and better” web data. It got me thinking about just how far analytics has come along since it first started. The tools we rely on today are a far cry from their “granddaddy”, 1993’s humble hit counter.
Ancient history: the hit counter
Analyzing web server logs
Search engines were a critical part of analytics before 1998, but web counters could not differentiate whether a visit was by a human or a bot, skewing statistics. WebSideStory, WebTrends, and other developers created web server logs, which could differentiate, human from bot visit, plus report information about user behavior like page views and session lengths.
Page tagging, cookies & packet sniffers
The concern for accuracy of logfile analysis and building analytics as an outsourced service has led to the invention of URL tagging, also known as page tagging. Another analyzer called the packet sniffer intercepted and logged traffic passing over a digital network to improve logfile analysis accuracy. This market grew to include embedded scripts on web pages, which gathered information the web server logs missed and assigned cookies to users to identify them during current and subsequent visits. Cookies are pieces of text assigned to each user that are sent by a web server to a web browser and are sent back by the visitor’s browser each time they visit that site again.
Analytics grows up
Software companies began packaging web services to provide a standardized offering to a growing market in the mid-1990s. What started as the optimization of hit count tracking and web server log analysis, gave rise to a market for sophisticated hybrid analytics services combining page tagging and logfile analysis and the growth of Google Analytics, Omniture, Unica. Coremetrics just to name a few of the household names in this space.
Predictive analytics: impossible or the future (now) of analytics?
Web analytics is about capturing and analyzing past visitor behavior on the site and using it to optimize various aspects of a visitor’s experience in the hope of improving conversions (subscriptions, sales, however we define them). But predictive analytics continues to be the hot topic. That’s because predicting user behavior is every marketer’s (and business owner’s) dream. Imagine using your data to forecast future visitor behavior and market trends, rather than just analyze the past… Ah… Wouldn’t that be grand. But even with our data collection becoming ever more sophisticated, there are so many variables and uncertainties, that it may not be possible at all. Given the history of analytics over the last 17+ years, I prefer to think that it’s simply not possible yet.