Disney Pulls Back Curtain On Ad Lab

New York -- Advertisers may know roughly how many people watch a given TV program. But how does television make viewers feel?

That's one of the big questions Walt Disney Co. wanted to answer with its advertising-research facility in Austin, Texas, which the company touted in a presentation here Tuesday to several dozen advertising executives.

Traditional TV measurement techniques are able to gauge what viewers "see and say" but don't measure "what they feel and think," said Peter Seymour, executive vice president of Disney Media Networks strategy and research. The Media Networks segment encompasses the Disney/ABC Television Group, ESPN, Disney Channel, ABC Family and other properties.

The Disney Media & Advertising Lab, announced last May, began conducting its first research projects in December 2008. To date it has fielded 16 studies and has 25 more in various stages.

The lab's goal is to decipher physical and emotional responses to television and other forms of video using proven scientific methods, said executive director Duane Varan, who is also director of the Interactive Television Research Institute at Murdoch University in Perth, Australia.

"Television is an intrinsically emotional experience," he said. "Fundamentally, we often can't articulate what is driving our behavior."

The Disney Media & Advertising Lab currently measures three primary types of physiological feedback: heart rate; skin conductivity (a measure of emotional response); and eye movement, to track where viewers are looking at the screen. The lab uses a goggle-based eye tracking system as well as a stereoscopic camera system that constructs a 3-D model of the viewer's head to track eye movements.

Disney chose Austin because it provides a good cross-section of the population. The company has assembled a pool of 3,000 people in the area who are statistically matched with the needs of each research project. Sample sizes for the tests range from 64 to 300 participants, Varan said.

The lab operates from 9 a.m. to 9 p.m., seven days a week, which means it can test 180 subjects per day and 65,000 in a year, according to Varan. The facility is identified as "The Media Panel" so as not to bias survey participants.

The lab's staff of 17 wears black lab coats. Noted Varan: "If someone is going to attach electrodes to your head, you want confidence that they know what they're doing."

So, what has Disney learned so far from the experiments? Mainly that some ad formats and approaches work better than others.

Seymour said one study demonstrated a notable lift in ad recall from product placement in ABC's Extreme Makeover: Home Edition. Two different groups were shown the same episode, including the same ad breaks with three 30-second Sears ads, except only one group saw the version with a 15-second segment of the show featuring Sears branding. Those who saw the in-program brand placement had a 27% increase in unaided recall and a 31% increase in purchase intent compared with the group that didn't see it.

Artie Bulgrin, ESPN's senior vice president of research and sales development, said a study looking at the effect of the network's Bottom Line ticker on ad performance showed there was no difference in ad recall between subjects who saw the Bottom Line during an ad break and those who didn't. Viewers spent only 12.6% of the time during the ad break looking at the ticker, the lab's testing found.

"Bottom Line presents no threat, and for ads that might fail to engage viewers, the Bottom Line provides a reason to stay on the screen," Bulgrin said.

What didn't work as well as Disney expected: Seymour said a study looking at ABC promotions on the lower third of the screen were only effective if they were preceded by a video promo. "For us, it was a little bit of a disappointing result," he said.

Mike Shaw, ABC's president of sales and marketing, added, "There may be times when the ‘A spot' is not the spot you want -- perhaps you've just come out of a drama and the human mind is not receptive to it."

As for how advertisers will be able to engage with the lab, Ed Erhardt, president of ESPN/ABC Sports customer sales and marketing, said the media company will evaluate requests on a case-by-case basis "with regard to our overall relationship."

One of the emerging technologies the Disney ad lab is looking at using, but has yet to validate, is facial coding. That measures the configuration of muscles on the human face to determine emotional engagement. However, according to Varan, no facial-coding software currently available is accurate enough. "Until it's validated it's magic, not science," he said.

Disney bookended the presentation with video clips featuring Dr. Marvin Candle, the scientist in the Dharma Initiative orientation films from ABC's Lost.

Candle, played by actor François Chau, deadpanned that the ad lab was conceived in 1927 by Walt Disney, "and it was to have been built on the moon."