To measure something, we have to first define it. What is Engagment? Engagement can be positive, negative, or neutral. Engagement can just be a promise to do something (e.g a contract) but one that need not translate to tangible outputs. Engagement can have any number of facets that the various gurus mentioned in their indices. Some folks say trust leads to engagement – I say, not necessarily. Nor does distrust leads to dis-engagement.

Anyway, the point my rant above is that I agree with you & Akin that it is better to measure eartly metrics like RFM (and a few others) than trying to measure an ethereal concept like engagement.

Regards,
Ned

I agree, traditional “engagement” metrics seem to not be very useful for my site, and recency/frequency are probably more appropriate. We have some tracking deficiencies right now that keep me from trusting those numbers from my tool (server logs with no analytics cookie, and authenticated visitors are semi-anonymous because of IP authentication), but as we remedy our tracking problems it’ll be interesting to see how frequency correlates with satisfaction.

Angie

Interestingly, I was just rewatching a webinar with Jim Novo and he had a very simple metric for engagement albeit much higher level. Namely, consistent with his Drilling Down book, he was emphasizing visit recency as an indicator of users’ degree of engagement. Meaning that the more recently a bucket of visitors visited, the more engaged they seem to be with the company. And when translated into conversions or sales the more recent ones would also be the ones – on average – to be the most likely to purchase, convert, etc.

I am probably butchering his message with my explanation but essentially it is a part of the well known RFM analysis.

Bottom-line, if measuring / predicting an individuals’ engagement isn’t working out, maybe one has to fall back on some higher level metrics such as RFM.

Thanks for a great story from the front lines!
Akin

Browser Wars!