Trolling time!

The customer journey, from becoming aware of a consumable good / experience, to engaging, to buying in, becoming loyal (or forming habit), and then disloyal as their trust is repeatedly betrayed and eroded, (or as habits drift) isn’t linear. In fact, it’s probably fraught with step-functions (tipping points in pop lit), and circuitous. Most disturbingly, modeling it all out would be some sort of euclidean n-space matrix. (Oh Noes!)

There was a Forrester report on engagement that portrayed the journey as being more like a series of pipes, with distinct outputs. For instance, it might be perfectly possible for somebody to become a champion for a brand without ever having purchased it – The Nintendo Wii during the supply gap, and the affinity that young people have for Mac’s without being able to afford them (yet!) are two solid examples.

There’s a dimension of ‘aspiration’ that I don’t think most of us are actively measuring.

For instance, let’s say a teenager clicks on a banner ad, on his 2002 Dell Computer, and goes to the Mac site. Loves what she sees. Mind you, Apple has already spent several dollars on everybody in an effort to get them aware of Mac’s, but there you go. She’s there now. She checks out the computer, browses models, sees the price, and runs downstairs to ask for one. Parents are currently reeling from a recession, so, 2008 dual processor mega wide-screen Mac is out of the question. Teenager *wants* a high end model. Teenager revisits the site from week to week, usually clicking on ads. She finally buys a Mac when she goes to university, through the site, student discount. w00t!

Currently, under most models, we’d say that her repeated visits to the site, but not converting, is somehow a failure of the site, when, in all reality, it’s a failure of her means to buy. The ads are possibly reinforcing her desire for a Mac though, and could be reinforcing the loyalty.

My point is – presently, with *most* direct attribution models, we’re missing most of the temporal bits of the equation. Previously, I posted that it’s conversion that’s the real dependent variable. And it is. But that doesn’t mean we should completely discount deferred conversion.

This aspiring teenager, in fact, will probably complain, at great length, to all who will listen, that she wants a Mac. There’s certainly an immediate social benefit that wouldn’t be captured through pure web analytics. We could, however, use social analytics to determine if the media is generating positive word of mouth (WOM) and emotional analytics to test for desire (Enter Carrabis stage right).

Our direct attribution models aren’t even all that clean in terms of direct attribution in many cases. To previous points made by Novo and Carrabis, there are precision/accuracy issues with certain measures. Social analytics is fraught with very high rates of error due, in no small part, to computability problems that are inherent in language itself (Enter Godel stage left). This is a real problem. While I believe that the specific choices of words we use reveal a great deal of insight into how we’re generally feeling, I think a lot gets lost in the semantic translation when we start stringing words together.

We also have issues in meme tracking.

We have baseline temporal models, in Novo RFM analysis. What’s really missing is a whole host of variables that are particularly difficult to track, and even harder to aggregate into a single database. Populating out that database is a critical piece of the puzzle. Then, we can have a more complete, indirect attribution model that is more holistic. I think that’s a very worthwhile goal for an organization.

May I posit that perhaps it would be important for major brands to run ‘aspirational’ campaigns, where the actual goal is to stoke aspiration, and then we could measure the results using the instruments that are appropriate to that goal? Maybe we should be running those instruments continuously, to track the effect of different events and campaigns, including the background radiation of the Internet on aspiration? Maybe this is the way to arrive at solid, tangible measures in absence of a complete indirect temporal attribution model.

I mean, the lack of a Grand Unified Theory doesn’t keep physicists from tinkering with the dynamics of shower curtains (ever wonder why the shower curtain always seems to fly at you when you’re in the shower? Oh yes, that question has been answered) or looking for neutrinos, does it?

One thought on “Aspiration, Temporal Analytics and Shower Curtains

  1. A big thank you to David Hamel for catching a font issue!

Comments are closed.