• by Cory Treffiletti , Featured Contributor, April 23, 2008

I tried. I really, really tried. Over the last few weeks I've created many opportunities for someone to give me a definition of the term "engagement" in its marketing sense, but alas -- I've yet to be given one. The term appears destined for use in our industry for many years to come, regardless of my relative defiance against it. Since the industry has spoken and you've all decided not to vote it off the island, then it occurs to me that maybe we should come up with a definition for it.

Last week at Ad:Tech, I began to think of a definition that would be easily understood, easily measured and standardized so that advertisers and marketers could apply it on an apples-to-apples basis for all formats of media. The term cannot be an online-centric term because that would be shortsighted. The term, if used properly, needs to be applied to all media vehicles and formats and needs to convey a single purpose and define a single component of said media.

Engagement, if used properly, describes the media vehicle, not the advertising. In television I've seen and heard reference to a Q-Score, or Quality Score. The Q-Score applies a value to the attentiveness of the audience watching a program. For more highly attentive, more emotionally invested programs the Q-Score can be high (think "LOST," "Heroes" and "Desperate Housewives"). A low Q-Score applies to a show that is viewed with a lower level of attentiveness. These are the types of shows that a viewer can become distracted from (think "Larry King Live," the evening news). Shows with a higher Q-Score are considered stronger for branding because the viewer is more involved. The shows with a lower Q-Score are considered better for direct-response advertisers, because a message that resonates well can create an immediate response.

The idea of a Q-Score is a good start to defining engagement, but it's not the only component of the measurement (plus I have to honest in saying that I don't know the methodology for that measurement). Attentiveness is but one of what I consider three possible components. The other two components are monthly repeat visitors and possibly a measure of growth or trends.

Attentiveness provides a measure of how deeply involved the audience is for a site, and a proxy for attentiveness might be time-spent. We would have to measure time-spent against categories and formats in order to be accurate. As an example, TV programs that are 45 minutes to an hour would inherently be more engaging than a Web site that offers a 10-minute (or so) experience, but a Web site with a 10-minute experience is infinitely more engaging than a search engine whose job is to provide a link to content. This measure of time spent might be one very strong component of engagement.

Repeat visits (on a monthly basis) provides a measure of "stickiness" or how often the user comes back, implying how invested the audience is in the site/magazine/newspaper/etc. For a Web site this measure implies a higher level of engagement than a magazine which only gets read once, whereas a site can be visited multiple times in a week or a day. A TV program gets watched at least once a week, and in the case of "American Idol," you get up to three in a week. That is a highly engaged program vs. some other shows.

Growth or trend data supplies a measure of where in their lifecycle that media format currently is. For example, Facebook would have a higher growth rate currently than MySpace, implying that users are more interested in Facebook right now than MySpace. It can be considered a measurement for a fad, or a highly temporary trend, but overall you can get an idea of whether a vehicle is in its prime -- or maybe it's "jumped the shark," both of which can add value.

These three measures can be used in tandem to provide some insights into the quality of the audience that's being exposed to a media vehicle. For a Web site these are very easy to gather. For more traditional media formats, the information is a little more static but just as important. Think of a TV Show like "LOST." The show is on four to five times per month, but some viewers watch the episodes multiple times to gather in all of the information. The show is 45 minutes long, so the time spent is much higher and if you average out these three measures you would likely see that a show such as "LOST" is more highly engaged than a number of Web sites. That's a hard point to argue, so the measurement makes sense to me.

Of course the other component to consider is the source of this data. For engagement to become a true metric, one of the third-party syndicated research sources would have to start tracking and offering this data. The formula for calculating the numbers would have to be approved by the industry and accepted by the marketers and the agencies who work with them. I hope to see this motivation in the coming months because, try as I might, I don't see the term going away.

What do you think of this proposed measurement? Am I missing something? Please respond on our message boards and let me know!