From TheFutureOf (16 Jul 08): Responses to Geertz, Papadakis and others, 5 Feb 08

Hello again,

I picked up the thread of this conversation at From TheFutureOf (10 Jul 08): Back into the fray and am planning on getting more involved in this blog again simply because some folks took the time to comment, therefore I owe them the honor of responding to their comments.

Yes, I know…I'm just like that. Anyway, here I'm picking things up at Dr. Geertz's February 5th, 2008 at 11:20 am comment (on TheFutureOf and sorry, I don't have a copy of it).


The Good Dr. Geertz, 5 Feb 08, 11:20am

Since Dr. Geertz published this comment he and I have had some rousing and wonderful phone and Skype conversations, one of which I extracted and published in My Easter Eggs Critiqued. Dr. Geertz was also kind enough to read through NextStage's patent (several times. Somebody buy that man a beer!) and lend his expertise to our internal discussions. It's wonderful when one finally learns the identity of one's advocates.

Dr. Geertz offers that I was suffering from some consternation in my remarks above. I thank him for the thought and suggest that I was biting my tongue until I knew we had the patent more than anything else.

And also and for the record, I do not object to anybody else's definition of engagement and their use of that definition as a description of some metric or tool they've developed, I only recognize that their definition is different from the one NextStage uses. I also believe that NextStage's definition is more realistic and calculable when it comes to determining future acts by a given (visitor) population across the broadest possible number of platforms throughout time.

Theo Papadakis, 5 Feb 08, 2:43pm


I read through your post on Avinash's blog a few times and applied both simplified logical calculus and Alexsander&Dunmel's Logical Calculus of Consciousness to it. I think there's a flaw in your original proposition. You write “If x is engaged with y, x is related to y.” This translates to something like “If y = f(x) then (x,y) are in some space S” which is true. Prior to that you write that what you're defining is a one way relationship and that's not valid. This inconsistency demonstrates itself later on in a logical analysis of the steps involved in the hypothetical demonstration you offer.

You write “two kinds of engagement with an object, positive and negative, by which I meant that someone can be engaged with an object.”

Hmm…I think I explained this in From TheFutureOf (10 Jul 08): Back into the fray, Online Engagement: What Exactly Is It?, Meet Online Engagement's Little Friend, Satisfaction and elsewhere. I'll also be following these up with The Money Is Where Engagement Meets Satisfaction Online

Engagement is the same because the same parts of the brain are active. You can be engaged by pain or pleasure. Your response to the pain or pleasure will (probably) be positive or negative and the parts of your brain telling you “Pay attention!” — those parts that dictate whether or not you're engaged — don't really care about your response, only that your attention is focused sufficiently such that the response can be effectively and adequately acted upon.

You then write “Engagement is not itself a psychological state but involves a mixture of rational beliefs and psychological states…”. I'll admit to having some challenges with this. I think my definitions of “psychological state”, “rational” and “beliefs” are inadequate to your use of those terms. Beliefs, for examples, stem from a part of the {C,B/e,M} matrix that doesn't rely on rationality.

I've written about the {C,B/e,M} matrix in several places such as Guest Blogger Joseph Carrabis Answers Dave Evans, CEO of Digital Voodoo's Question About Male Executives Wielding Social Media Influence on Par with Female Executives, Responding to Christopher Berry's Vexing Problem, Part 3 post, Responding to Christopher Berry's “A Vexing Problem, Part 4” Post, Part 2, From TheFutureOf (27 Feb 08): Now onto Eric's 'If I'm translating you correct, …' comment.

and several other places. I can also offer a bibliography of how the {C,B/e,M} is derived for those with an interest.

A simpler version of the {C,B/e,M} matrix is described in Reading Virtual Minds and deals with the different ego states of Core, Identity and Personality. Where ever they come from, beliefs may be informed by data but they are not bound to rational thinking. They are, as described in my exchanges with Christopher Berry, based on anecdotal thinking and subject to all the pitfalls inherent therein. You have a good primer on this material in Schacter and Scarry's Memory, Brain, and Belief

Likewise “rational” is not an adjective I'd use with “beliefs”. “Psychological states” I recognize as “a mental condition in which the qualities of a state are relatively constant even though the state itself may be dynamic”. Some worthy albeit diverse reads on this subject include Tomasello's The Cultural Origins of Human Cognition, Gazzaniga's Mind Matters: How the Mind&Brain Interact to Create Our Conscious Lives and Allison's Minds in Many Pieces (sorry, couldn't find those last two online).

You also write about the “degree of engagement” and tie your definition to “psychological investment/involvement”. I'm quite sure I don't understand the use of these terms in these contexts. This is not to indicate they are invalid, only that I don't understand them.

Lack of understanding, however, is much the central problem that needs to be addressed in these discussions. Any metric is meaningless unless the language of what is being measured is specific. It might make good business sense that some group has a proprietary definition but when that definition no longer applies the business suffers. It's probably better (opinion warning) that a common definition be used and that businesses work at being more accurate in providing a metric based on that common definition (just as a point, NextStage uses common definitions that anybody can use. It's our methods and technology that are proprietary, not our definitions).

If I understand your hypothesis correctly, you suggest that your definition of engagement can't be measured, only inferred. I don't think I'll argue that point, only question it's utility. Breaking down the logic I come up with

a) Hypothesize some phenomena
b) Define the phenomena such that it can't be measured using standard and readily accessible tools
c) Take measurements using standard and readily accessible tools
d) Develop a formula that takes the measurements from the above
e) Claim that what is measured, via the formula, is the phenomena.

There are some logic holes here (to me, anyway) and I won't go through them in detail. I will offer that there must be an unbroken chain of physical connection between how something is measured and what something is in order for that measurement — hence the resulting metric — to be valid.

I'm happy to continue this discussion if there's interest. I do want to point out that the lack of precision in both definition and logic will probably be impedimentory. Do also note that I don't recognize problems with “measuring degree of engagement” based on the definitions I use and the measurement methods applied (the three “real questions” in this thread's original post).

Eric Peterson, 5 Feb 08, 8:28pm

“Save asking every person who comes to a web site are you engaged? (which I would assert is A) impractical and B) just as imprecise as my calculation, if not more so!), how would you propose we ground truth engagement and test the hypothesis?”

I can't speak for Dr. Geertz and this is where that nap-of-the-earth flying thing comes in, I guess. I've often described Evolution Technology (ET) as doing exactly that, asking every person to a website “Are you …?” and then responding via whatever business rules are in place. Hence this is not impractical. Is it imprecise? That depends on how much precision you'd like. In more tests than I care to remember NextStage's ET averaged 83% accuracy predicting outcomes (what people would do, when they would do it, etc.).

Re “satisfaction”: Yes, agreed. This is something I mentioned during one session at eMetrics SF 08 and also in From TheFutureOf (10 Jul 08): Back into the fray and Meet Online Engagement's Little Friend, Satisfaction. I mentioned a future column post on this subject I'll share some highlights here.

Engagement and Satisfaction as an XY PlotFirst and for explanatory purposes, I'll shade in the figure in From TheFutureOf (10 Jul 08): Back into the fray so that it looks like the one on the right.

Those shaded areas actually do have meaning. People may have heard me mention that ET makes decisions and offers suggestions based on solid probabilities. The combination of those shaded areas (and their extensions) creates a bell shape. Imagine that bell filled with metal. Now you have a solid bell. Imagine every atom in that solid bell representing a probability that something will or won't occur, that a visitor will do this or that, will respond one way or another, will think this way or some other way. The accuracy of the prediction — the likelihood or probability that something will occur — is based on the atom's height within the bell and its distance from the bell's surface. Things closest to the center of the bell are more likely to occur than things at its edge.

Thus something midway up the edge of the bell is less likely to occur than something at the bottom of the bell but dead center.

Hey, this is my world. I've learned to live in it and I'm not claiming anybody else has to or should.

Click for larger imageWhat happens when you map those quadrants onto a standard Engagement chart is something like on the right. The money (if you will) is in the periodicity. The periodicity can depend on several things, most of which are business dependent. What the periodicity gives you (via linking to some standard (at least what I think are standard) web analytics values) is a near surgical ability to recreate optimal satisfaction-engagement periods at will.

Thus ends the 5 Feb 08 comments

Again, whoosh! At least I got a column out of this. Next time I'll start with Dr. Geertz's 6 Feb 08, 11:56am comment.

And thanks for everyone's patience.

17 thoughts on “From TheFutureOf (16 Jul 08): Responses to Geertz, Papadakis and others, 5 Feb 08

  1. Dude I stayed up all night reading these posts. I am almost embarrassed to admit that I can follow maybe half but you've got me hooked.

  2. I have more than 12 nich advertising markets I do analytics for in the midwest, search engine optimization, pay per click adwords, omniture and social media setup. I'm commenting here but everything in these posts is so true.

  3. Wow. 12 years in the “online” analytics business and now I see this? I've been going through your posts over the holidays and am stopping here to comment. Thanks for putting into words what is common but unspoken knowledge here in KCMO.

    1. Hello to both you and Ms. Acasi, Mr. Land. From looking at the time stamps on your posts it seems you were talking before posting. Ms. Acasi's post made it into the system before yours.
      In any case, my thanks to you both for reading and commenting.

  4. I saw that your blog made the top tweets site 100tweets of whatever its called so I came by to see what it's all about. Way over my head but congrats anyway.

  5. Same as Floris. Came by to see what's what. Congratulations.

  6. I came by because of that top tweets thing. No idea what this means but good for you.

  7. Me, too. That 100 tweet thing.

  8. That 100 tweets thing brought me here. Congratulations on the recognition. What's this blog about?

    1. Greetings Sheppard and Paladin,
      To the latter, you're a fan of mid-80s television SF and/or Harlan Ellison (my guess based on your chosen name)?
      Whatever brought you here, welcome.

  9. […] “Credibility deals with facts and Believability deals with emotions and desires. Literally we’re dealing with the differences between scientific thinking (Credibility) and anecdotal thinking (Believability). Again, this is something I wrote about in Responding to Christopher Berry’s “A Vexing Problem, Part 4″ Post, Part 1, My Easter Eggs Critiqued, Defining Engagement (Again? Oh, Lordy!) and Exploring the Holes in Flawed Logic and From TheFutureOf (16 Jul 08): Responses to Geertz, Papadakis and others, 5 Feb 08. […]

Comments are closed.