The Analytics Assessment: Gary Angel Goes One-on-One with Web Analytics Guru Eric Peterson


Isn’t that quote about “Giants” from Newton (who would have squashed just about anybody I’m thinking)? Education isn’t all one-way “push”, of course. What should people bring to a webinar like this — is there any work it makes sense to put in before hand. Or is this like a first piano lesson — you can’t do much until you know a few basics? And you’ve heard the old quote that you don’t really understand something until you explain it…are there things we should be learning from newcomers?

E. Peterson: My gut feel for the type of folks who will attend this American Marketing Association/Aquent sponsored event is that they already have some experience with web analytics … they’re doing it, they’ve done it, they know they need to do it … but they’re not 100% satisfied with the results to date. My hope is that by clearly outlining these “Ten Steps to Success” that people will fill in the gaps between what they’re doing and what they need to be doing.

So some attendees with be doing two of the ten things I suggest, and others will be doing eight. Everyone should learn something, regardless of their experience with web analytics to date (okay, except maybe you Gary. You specifically might not benefit from the presentation)

I want to go back to process before we wrap up, but I had one last question about getting started in web analytics. In programming, there are languages that seem to be more appropriate for beginners (like Basic) and languages that are better for experts (like C++) — but as programming languages have matured those differences have seemed to matter less and less. The languages became easier to use and — in many ways — gained in power too. In your view, where’s web analytics in that scale. Are the best and most powerful tools also the best tools for beginners? Or does it make sense to start with a certain kind of tool?

E. Peterson: I think it’s more complex than just a line between “beginner” and “expert user” when you’re choosing a tool. Because nobody goes to college to study web analytics, every “beginner” brings a wealth of experience and understanding to the table when they get into web analytics. So I’ve seen some people come into web analytics and immediately push the limits of their application. Conversely, I’ve seen some “expert users” massively underutilize even the most basic tools.

It’s too easy to say, “Oh, you’re a beginner so you should use application X but avoid application Y” or “You’re an expert, you need application Z” because, in my experience (and this is cliché) it’s not the tool, it’s how you use it.

I think that makes a lot of sense. Sometimes these clichés are just true. I want to back to that process discussion before we wrap up. You’ve been a big exponent of the idea of building a process around web analytics. That seems to me obviously right and necessary. How much of that is organizational and how much is for an individual practitioner — in other words, is process the concern for a WA manager or an Analyst — or do they both have to be process aware?

E. Peterson: Megan Burns at Forrester Research says it best when she astutely points out that you’re not really building a process around web analytics, you’re integrating web analytics into existing processes. If you think about it this way, the Web Analytics Business Process is an organizational concern that touches different people in different ways.

The analyst has a specific and special interest in the processes because he or she is on the hook to produce high-quality analysis of the data. But the data collection is often owned by IT, who has a different interest in web analytics. And, ideally, even the CEO has an interest in web analytics.

In my humble opinion, everyone in an organization trying to be successful with web analytics has to be process aware. This awareness cuts down on the “oh, we forgot to tag that campaign” and “I thought we’d already validated this data” conversations that severely undermine the organization’s ability to use web analytics to drive business success.

I’ve always been a little distrustful of very formal process methods like Six Sigma — that seem to be rooted in the culture and success of one very influential company and often feel faintly ludicrous when transported elsewhere. In data analysis, I saw whole Business Intelligence companies built around systems that were really only appropriate for P&G. How much of a process is necessarily going to be industry (even company) specific and how much can we hope to generalize?

E. Peterson: I personally believe that there are a handful of generalized “Web Analytics Business Processes” that are easily described that are applicable across many (if not all) businesses. These are the core things like “Assign Ownership of Web Analytics” and “Plan for Experimentation” that are business-independent but critical to success with web analytics.

That said, the specific instantiation of each process is unique to every business. I’ve worked with a lot of companies in my nearly 10 years in this industry (geez, now I feel old!) and the one thing I’ve learned is that every company is different. So trying to force every company to approach these processes in the same way would be a mistake.

The idea is to get companies to slow down and recognize that these processes exist and need to be carefully considered. Again, in my humble opinion, one of the things that has repeatedly hurt our industry is the ad hoc, one-off, last minute attention that web analytics often gets. Measurement falls out of project plans and companies lose their ability to execute vis-a-vis web analytics when a single person leaves the company.

Process prevents that, or at least it is designed to.

G. Angel: Great answer. And, obviously, web analytics process is nothing like as formal as Six Sigma. We don’t seem to have belts to hold our pants up, much less “black belts.” I’ve used up a goodly chunk of your time – thanks so much! Any last words or thoughts?

E. Peterson: Last thoughts? Gary this has been a great interview! Some people have commented that our tête-ÃÂ-tête over visitor engagement and “the myth of actionability” has been a tad removed from the day to day concerns of most practitioners but I disagree. I think it’s tremendously important to talk about where we are today and where we want to be. You’re leading the way in that regard and I appreciate it.

Thanks again for having me over (as it were) …

Thanks. I feel the same way. Not that what we’ve been talking about in those posts IS a day-to-day concern because it isn’t. But because sometimes you lose sight of what matters when all you think about is the day to day stuff. Anyway – good luck with the webinar and have a great weekend!


Please enter your comment!
Please enter your name here