Chasing Paid Engagement

Posted by Tom Hespos on November 3rd, 2010 at 7:49 pm

By now, many interactive marketers are familiar with the notion that engagement is something that needs to be defined on a custom basis.  Conversations have shifted from trying to define engagement in some sort of industry-standard way to talking about how smart companies approach defining engagement themselves.  A morning panel emceed by Mike Darviche of Networked Insights showcased a few ways in which we all might start to think about engagement.

Before showing the room his framework for thinking about engagement, John Lowell of Starcom MediaVest Group showed the room the complexity of the variables involved by talking about Netflix and their movie-rating system for a bit.  Whereas rating a movie on a 1-to-5 scale gives others considering watching a film an idea of whether you loved it or hated it, the model is one dimensional.  Adding additional dimensions to the rating scale by introducing context ("What's a great movie I watched with my family?" "What is a great movie to see in the theater instead of at home?") can make a ratings system more valuable by getting closer to what the user really is in the mood to see.

"People are complex and weird, and five-star ratings don't tell the story of us," he said.

Lowell then introduced a framework for thinking about engagement that dove deeply into  understanding audience motives. Urging us to think about all the dimensions involved, he talked about the alignment between Target's commercials during the final season of "Lost."  The clever way those commercials integrated things from the show like the Smoke Monster with products that could be found at Target (e.g. - a smoke detector) was aligned with the motives of the viewer in expressing their engagement with the show.

If Lowell's presentation was about how to conceptually wrap one's mind around engagement, Jason Krebs' presentation was about the rubber meeting the road.  Krebs, EVP at ScanScout, brought a Jeep case study with him.  In showcasing three different types of ads that ScanScout ran for Jeep, Krebs also gave the crowd something that seemed to be counter-intuitive.  The industry argues about the notion of whether or not videos can engage the user long enough for a 30-second spot to get across, but Krebs showed how some ScanScout units were able to keep Jeep's target engaged for 110 seconds.  Featuring microsite-like content that allows interested viewers to request more video, ScanScout's units can be priced on the basis of how many people choose to interact with them, which presents an interesting Cost Per Engagement model for advertisers to take advantage of.

Heather Dougherty of Experian's Marketing Services division lent a different perspective to the panel - One that showcased marketers' missed opportunities to engage users beyond content.  Leveraging Hitwise, the company's behavioral data play, Dougherty showed some compelling search data that looked at how users continued to look for content based on some of their favorite reality TV shows.  One example centered on Dancing With The Stars lent some dimension to viewer search behavior as they interacted with the show.  This revealed some interesting potential opportunities for ABC, as an independent site was tapping into that search behavior to build its own conversation around Dancing With The Stars.   Could this be an opportunity for ABC to partner?  Then again, ABC wouldn't know about the opportunity if it didn't dive into viewer search behavior.

We're far from an industry standard for engagement, and that's probably a good thing.  One big takeaway from this session was that engagement can be thought about in many different strategic and tactical ways, but they all come back to a desired behavior on the part of the people advertisers are engaging.

2 Responses to “Chasing Paid Engagement”

  1. Great post Tom! I think you definitely captured the main points from the presentation. To add to your post, I have some additional thoughts around Mike Darviche opening presentation.

    I found it interesting when he highlighted the top shows based on Nielsen Ratings compared to the top talked about shows across the social web. In one example he had House at #9 for watched, but #1 for social conversation. In asking the question if the two correlate, meaning if a show drops in the ratings does the conversation also drop - his answer was no. A perfect example of this is Family Guy. A show that Fox cancelled but brought back after the online audience went crazy and bought the record setting amount of DVD copies. This take way was key to how brands need to ensure they have the right mix of on and offline actions to drive engagement.

    I also found the study on niche properties and the time of year, played a big role in engagement. The example of a brand advertising on the FoodNetwork.com around Thanksgiving translated to the highest amount of visits for the month was interesting. This example took the standard tactic of "content relevance" and added a layer of "holiday relevance".

  2. Bonsoir ! J'ai noté que votre blog est bien référencé mais des optimisation supplémentaires sont possibles. Avez vous démarché un consultant seo sem ? Sinon, contactez-moi ;)