Web Analytics Symposium Reivew

9 Flares Twitter 0 Facebook 0 LinkedIn 0 Google+ 9 9 Flares ×

This week I had the opportunity to attend the Web Analytics Association (WAA) Symposium in Santa Monica.  Overall it was a well put together event, and while there weren’t a lot of actionable tactics to take away, there was an abundant amount of strategy level discussion.   Here’s some of what I took away:


Attribution – the hottest topic on everyone’s lips was attribution.  I met several vendors working on products to help solve the multi-touch attribution issue.  In other words, if you made contact with a customer through multiple advertising channels, which spending account do you attribute the sale to?

Sentiment – While it seems like behavioral analysis is well understood and powerful tools already well know, there was a lot of interest on how to quantify the fuzzier topic of social sentiment.  It was noted that Radian6 (recently acquired by SalesForce.com) seems to be the leader in looking at this question.

Iterative Optimization – It was interesting to see the analytics community seems to feel a lot of ownership over the topics of A/B and MVT testing.  I suppose it makes sense, since the optimization of an application is ultimately accomplished with analytics tools. I didn’t expect to see this however.

Setup v Analysis – One speaker gave a talk on what he called “setupland versus action land”.  He described how after years of work as an analytics consultant, he watched the majority of his clients get stuck on implementing analytics packages (tagging, etc), and very little time actually analyzing the data.  It seemed many in the community related to this challenge.

Tag Management – There were three different vendors at the conference offering what they called a tag management system. If you embed a single javascript tag into your page, you can use their system to “walk the DOM” of each page, and bind analytics events to various interface interaction points.  Its a great solution because a company can spend a lot of time and money tagging their application and becoming higher dependent on a single vendor. By using a TMS, swapping out vendors is much easier.  it also makes it easier for the marketing dept to easily embed javascript tags without needing to burden the IT department, a common challenge in large organizations.

Goal-Driven Strategy – This seems almost obvious but seems to be a consistent problem as many people touched upon the issue.  The goals and KPIs of a project must be clearly defined up-front, in order to direct or measure the success of the initiative. As annoying as it can be, Stakeholders need to be challenged to provide well defined objectives and priorities before implementation begins.

Profile of Attendees – The vast majority of those I met at the Symposium, were in-house analysts at large companies. There were certain companies that seemed to have a larger presence that others, such as Kelley Blue Book and Experian.  I met a few consultants, people from local universities and even local marketing companies.  Interestingly, the vast majority of the attendees use some form or another of Omniture/Adobe tools. I heard of very few people (if any) using Google Analytics at the Enterprise level.  I also didn’t hear much mention of WebTrends or CoreMetrics this time.

Maturing Discipline – Finally, it was interesting to observe that what is still a quickly evolving area, is already beginning to look mature.  It was apparent in many ways. Large companies have acquired smaller ones and have entered niche markets.  For example, Adobe purchased Omniture, and now offers even a Tag Management System, competing with the niche guys.  I also heard rumors that IBM is working on collecting and building a massive play in the analytics space.  I heard also that in many organizations, analytics is now understood as a KPI for the business and thus more analytics heads are reporting directly in to the CFO, not the CTO.

All in all, it was a great meeting and good to see what some of the leading West Coast analytics minds are thinking about and working on.  I think there are a number of lessons that can be applied to smaller organizations and application development cycles as well.