Advise and Consent

Comments on Web Measurement Organization and Bringing SEM In-House

There have been several comments on recent posts and each has at least partially changed my mind. Marshall Sponder and Jacques Warren had several interesting points regarding my post on locating a web measurement department. And Chris Spiek wrote both a comment and a detailed post on the whole issue of taking SEM in-house.

I’ll start with Marshall’s comments, because they are easiest to consider. Marshall makes two points concerning my discussion of where to put SEM Measurement that I pretty much agree with. First, he argues that my example of a case where web measurement is organized to satisfy a single important executive in a company isn’t a good paradigm. He’s right, of course. I didn’t really mean to suggest it was.

Extremely dynamic and powerful individuals nearly always strain and distort company org charts – and this phenomenon is general – having nothing to do with web analytics per se (or even corporations – it happens in any sort of organized activity). I’m not sure how avoidable this problem is – or even how much of a problem it is. Companies should be aware that there is a price and, potentially, a benefit to be had from these distortions. There are times when it is appropriate to "bet the company" and let organizational lines go to pot (I always think of Tracy Kidder’s Soul of a New Machine as an almost novelistic case study in just that question). There are times when it isn’t. Deciding when such bets are smart has nothing to do with web analytics and is well beyond the plausible scope of a web analytics geek like me running a 15 person company!

Marshall’s second point is more interesting and, I think, more significant. He points out that just because a company has a Strategy Office or CSO doesn’t mean that’s where strategy is actually done. Not only is this undeniably true, but it points to a deeper truth. A big part of placing a web measurement department within your organization is understanding where having it can best influence, shape and enforce your vision for the web. There’s no reason to believe that place will be the same for every company.

Which brings me to Jacques Warren’s comment. It was Jacques who – more than anyone else – convinced me to add comments to my site. And his example here is a good example of the reason why. I think Jacques is probably pretty much on target that Marketing IS where web measurement will end up. He’s also right that putting web analytics too high up in an organization risks making it irrelevant. I am, though, rather less sanguine about the transformative nature of web analytics and marketing. We’ll see!

Finally, there is Chris Spiek’s take on the whole taking SEM in-house debate. It’s worth reading his actual post -http://www.awecommwebstrategies.com/blog/ - (which is very good) and not just his comments on this. I think Chris makes his case about as well as it can be made – and puts the emphasis in all the right places. I’m not sure I end up thinking he’s made a case against anyone taking SEM in-house, but he’s at least made a pretty good case for trying someone like him (or AWS) before you do. Which is a pretty darn good case to make, all things considered! Since he brings this issue home to me by using web analytics, I’ll say in reply that I’d be happy to have made an equally good case for using Semphonic instead of building a web measurement department in-house.