Erwin Ephron

Engagement Explained

The Confusion is Engagement is Many Different Things.

"Engagement" is more than a search for accountability. It is a cry for help. Advertising has become more fragmented, more costly, and less effective. Many advertisers hope engagement will be the tool for making things better. A system for improving consumer response.

But doing is always more difficult than saying, so what do we mean by engagement and how might we measure it?

MEASURING ENGAGEMENT

First, what are we talking about? Idea-words like "engagement" as distinct from thing-words like "pencil" are tricky. You can't see, hear, smell or touch engagement or chew on its eraser. Engagement exists only as an abstraction.

But we're used to working with abstractions. Our favorite is "TV Audience." Audience is not a thing. It exists in space and time only as a Nielsen estimate tied to related, measurable consumer behavior: "People sitting in the room with a live TV set, who have pushed a button to indicate that they are viewing." We don't call them "Nielsen button-pushers". We call them "audience."

What measuring audience tells us about measuring engagement is we need to redefine it as something that helps us to predict viewer response and can be measured. Otherwise engagement is just a feel-good word without substance.

But before we even begin measuring, there's an important caveat. Engagement with a program is not the same as engagement with the commercials a program carries. Viewers make that distinction so we must also.

For example, data show that in DVR households the more engaging programs - higher rated sit coms, dramas and movies - are most frequently time shifted and fast-forwarded through commercials. So the most engaging programs on TV may well carry the least engaged-with commercials. To avoid this confusion we need to focus on engagement with the ad message carried by the program, not on the program itself.

ENGAGEMENT IS NOT ONE THING

A second problem is we tend to think of Engagement as a single thing: A consumer state which results in greater likelihood of response to advertising. Engagement as a measurement is not a single thing. In TV viewing it is the sum of all measurable variables that significantly affect the probability of viewer response to the ad message.

These variables include:
Size of the unit. Because recall studies tell us that a 30-second commercial is more likely to produce a response than a 15 is.

Clutter. Because a nine message pod will lose more viewers for the average commercial than a three message pod.

Situation. Because the number of persons in the room, the location of set, the time of the telecast, multi-tasking, etc. will affect who sees and gets involved with the commercial.

And finally, Relevance. Because what the commercial is about helps to determine viewer interest and response.

A HEAD TWISTER

This four-variable formulation of engagement is a head twister. It says that much of engagement is not just about content. It's also about the mechanics of TV message delivery.

Engagement measures must first adjust our audience data for the likelihood that a message Nielsen reports as "exposed" will actually be seen by the viewer, because commercial inattention and avoidance is often the reason viewers do not engage. For example, an engaging commercial (our definition) proceeded by six loser messages is still less likely to be seen and responded to, no matter how engaging.

Relevance comes closest to the feel of the word "engagement" in measuring the consumer connection. It is defined as the closeness of fit between the characteristics of the viewers to the program carrying the advertising and those of the ideal prospect the advertising is trying to reach. Said another way, relevance is successful targeting as experienced by the consumer.

The supremely useful thing about this set of four engagement variables is three are being measured and are in the TV database right now. Only relevance requires a short trip to MRI. And some substantial research is needed to dimension the fairly obvious effects of shorter messages, clutter, situation and better targeting.

HOW DO WE USE THE DATA?

Now that we have defined what to measure and suggested how we might measure it, there is a second challenge. How do we use the data?

It's over-reaching to think of engagement as a substitute for frequency as in "Reach x Engagement = Response" and present this as a better media planning model. It is more realistic to use engagement adjustments to net-down current TV ad exposures to those more likely to obtain a response. This will create a new, more accountable TV currency.

Today "Engagement" is more than a ring, an appointment, a match or a battle. It is a cry for help. It may not be simple, but help is on the way.

- November 21, 2005 -