I proposed to my wife on a Monday evening, and by Wednesday our apartment was littered with these magazines.  Because once she became, you know, engaged, she simply had to know "The right words for your invitation," "Ten ways to serve little hot dogs in blankets," "Helping him choose the right cummerbund," and all sorts of other vital stuff.  In fact, I would say that these magazines (and their Web sites) have the most engaged audiences in media.

Don't worry, I won't keep this conceit up all column.  I just wanted to ease you into yet another essay on engagement, that ephemeral quality that all of us in the media business pursue doggedly, even as none of us really, truly, know exactly what it is (save for the bridal magazines -- and hey, my veil is off to them.)

What IS engagement, exactly?  The good people at the Advertising Research Foundation defined it as "turning on a prospect to a brand idea," to which my friend and colleague Erwin Ephron replied that perhaps we could measure it with a "blush-o-meter." (Followed immediately by a call for the MRC to audit the blush-o-meter before it goes live.)

I'm here to tell you I don't actually know what engagement is -- at least not precisely.  In fact, I was on a panel at the IAB Leadership Forum on Performance Marketing last week, and there my co-panelists suggested that engagement would mean different things to different brands and verticals.  What I do know is that historically there were two dimensions to media evaluation: reach (how many different people) and frequency (how much, or how often, did those people consume the media vehicle or ad campaign?)  Media was two-dimensional; reach X frequency told you all you needed to know.

Today, though, we seem to have accepted that not all media exposures are created equal.  Engagement is the term we have aligned around to describe what's going on in those exposures that are more equal, or valuable, than others.  So media math is fundamentally changing: how many; how much; and now, how good.  For advertisers it is no longer sufficient to get your ads in front of the target audience with effective frequency; now, during those exposures, we expect some cognitive thing to happen; some synaptic firing, some ineffable spark, some tiny but profound connection between consumer and content. 

As Web analysts attempt to operationalize the concept of engagement into a meaningful metric (or suite of metrics) for tracking and evaluating the performance of a Web site, they start to talk about things like visits, time spent, clicks, returns, conversions.  All of this serves to remind us that, at its core, the engagement of a Web site is made up of the behaviors of people, the sum total of the individual experiences that consumers are having with the Web site each day.  Those experiences will run the gamut from the sublime to the ridiculous. For the Web publisher seeking to build a more "engaging" site, the goal is to identify the site's target audience, and then strive to assure that the consumers comprising the target are having the sublime experiences.

A lot of companies in the online metrics space (including, well, my own) are trying to define an engagement metric to beat all engagement metrics.  One piece of work that seems to be getting both peer vetting and traction is "the Peterson Calculation," developed by Eric Peterson and already discussed in this space by Kevin Mannion.  Peterson defines engagement as "an estimate of the degree and depth of visitor interaction on the site against a clearly defined set of goals." (Italics added.)  I like that definition because it acknowledges that engagement itself is flexible (depending on what your specific goals are), while forcing it to be profoundly measurable, since this definition requires establishing clearly crafted goals in the first place. (The researcher in me finds this more useful than "turning on a prospect to a brand idea.")  You can find the guts of the Peterson Calculation about halfway into this pdf.  

Now, of course this calculation is for Web sites, not ads; there is a whole 'nuther body of work that needs to be done to tell us what we need to know about the relationship between engagement of media vehicles and the ads therein. We generally assume that the consumer engaged by media vehicle content will be more likely to engage with ads within that content; personally I think that all depends on the message and execution of the ad, and the interests and needs of the individual consumer.  Indeed, the same consumer at the same precise Web page might be a highly engaged viewer of one ad and wholly indifferent to another.

But I do think calculations like Peterson's will prove useful, especially to Web operators who want to track their own performance over time with respect to engagement of target consumers.  Maybe Peterson's calculation is the right one for your Web site.  Maybe you need to tweak it a little, or invent your own.  The important thing is that, in order to strive for those more meaningful, higher-quality consumer interactions -- the stuff engagement is made of -- you need to (1) conceptualize what experiences you want your visitors to have, and (2) what component parts make up those experiences for your visitors; then you can (3) set goals and track your performance accordingly. 

You may not get an audience as consistently engaged as that of Modern Bride.  But you'll do a lot better than Today's Bachelor.

This commentary is insightful. I recommend it to others.