The Web Analytics Revolution


dodgeball2.jpgThe Web analytics industry has come on in leaps and bounds over the last few years. The rise in prominence of relatively easily deployable tag based systems has catapulted the industry into the mainstream and everyone from the smallest mom & pop operation, right up to giants like Comcast & QUALCOMM (both clients of mine), have embraced the discipline. The extent to which their analytics programs integrate within their existing business of course differ greatly, but whether it’s Joe’s Crab Shack down the road or a global corporation with thousands of employees, the ultimate goal remains the same: performance insight. Each presents different challenges, but the objective remains the same.

Why is this a revolution though? A lofty claim, for sure. There are four key differences in Web analytics today vs. a few years ago, all of which are significant contributing factors:

It’s so easy to get at your data now. Whether through custom dashboards, the interface itself or your own custom reports populated using your vendor’s API, it’s never been easier to quickly get the information you need in the right format. Huge plus point.

KPIs vs. FGRs (Feel-Good-Reports)
In years gone by, execs would get a 100 page report on their desk that made them feel good (hence feel-good-reports, with apologies to whoever coined the term). Nowadays, the analytics industry is focusing on Key Performance Indicators: specific metrics relative to that particular business, the monitoring of which provides meaningful performance insight. Doesn’t that alone make you feel good?

Today’s tools integrate beautifully with complex Web technologies. Want to track a fully featured flash movie? No problem. AJAX site? That’s fine too. Being able to fully understand user behavior across your entire site is critical.

Analytics is mainstream news nowadays (for example: note the recent news stories around Neilson’s adjustments around “time spent” vs. page views.) Five years ago analytics advocates would frequently encounter blank faces as they evangelized. Nowadays, analytics is so prominent (in part because of the reasons outlined above); it’s a much easier sell.

Wanting to better understand Web performance is nothing new. From the early days of log files (WebTrends) vs. page tagging (WebSideStory), through today where we see analytics working its way into online business processes, the underlying driver of wanting to better understand user behavior in unchanged.  The key difference nowadays is accessibility. With very little overhead almost anyone can deploy a sophisticated analytics solution and quickly start collecting real time user behavior data.

Free packages like Google’s offerings are perfect for non-profits or smaller sites looking for relatively basic functionality. Higher end solutions, like Site Catalyst from Omniture, offer a much more robust feature set (robust is underplaying it, actually, feature rich does not adequately describe what Omniture is doing: further disclosure, Omniture is a partner of ours). In short, there’s something for everyone out there. What’s right for one client may not be right for another, but all requirements are catered to if you’re prepared to do a bit of looking around.

The real power of the current crop of analytics packages (Site Catalyst, HBX, Google Analytics, WebTrends, Index Tools and so on) lies in their versatility and ability to deliver value to so many different types of business. You don’t need to invest a huge amount to get an analytics program up and running, although (of course) the more advanced your requirements, the greater the setup and maintenance cost. For any site with a commerce element, analytics should be a no-brainer, but there are less obvious (though no less important) applications also.

Brand sites wanting to drive users to particular types of content, particular conversion events (email signups, registrations etc) are all analytics contenders. A solid analytics strategy will help you optimize your efforts across all these areas. Although the investment can be modest, it is also important to bear in mind that for your analytics program to be a success, just as with SEO, you need to integrate analytics into your overall process. It’s an ongoing initiative, not a onetime deal. Any and all changes to your site need to be undertaken with your analytics implementation in mind (of all implementation problems, mistagged pages is one of the most common. Incorrect page names or account numbers compromise the integrity of the data that comes out the other end) – this is doubly true when you have multiple folks working on the same site.

What does this mean for us marketers though?  One of the most powerful developments in Web analytics over the last few years, and one of my ‘revolution catalysts’ outlined above is simply how accessible the data has become. Having that real-time window on performance gives you an immediacy of feedback that was simply not possible a few years ago. You can launch an initiative and you get quantitative data in front of you right away.  In short, Web analytics can help you take the guesswork out of your marketing efforts.

If your organization doesn’t have an analytics program in place, why not consider looking at setting one up? If you do have a program, are you getting the most out of it? Are your stakeholders getting the information and analysis they need, and most importantly is the analysis that your analytics program drives helping you and your stakeholders make business decisions?  There’s a dizzying array of data available out there. Are you getting the most out of it?


  1. I wholeheartedly agree. Even know people use the wrong KPI’s to gauge progress. Some people use time on site as a metric, which can be altered depending how aggressive you drive your traffic to your website. Web analytics should be married with scent trails and predetermined funnel paths. You need to first figure out how you would like a visitor to flow through your website and then figure out the obstacles standing in their way.


Please enter your comment!
Please enter your name here