Agency creatives and other critics have long suspected copy tests are often wrong and favor colorless, unlikable ads. But the new attacks on testing go a step further by analyzing why, and the criticism is coming not just from agencies but also from clients and testing services.
"It is time ... to update our outdated copy-testing techniques," said Ken Kaess, president-CEO of Omnicom Group's DDB Worldwide, and chairman of the American Association of Advertising Agencies at the association's recent management conference in New Orleans. "Advertising works best by changing or reinforcing qualitative perceptions over a period of time."
Bearing the brunt of criticism is the oldest and one of the most widely used testing techniques-recall-which focuses on whether consumers remember a product or brand after being exposed to an ad or whether the ad persuades people to buy a product.
"What we hear from the advertising agencies working on our brands is that recall [testing] actually produces results that are counterintuitive," said John Kastenholz, VP-consumer and market insights of Unilever Home and Personal Care, in a speech presented at the Advertising Research Federation's convention in April. "There's a feeling, particularly among creative directors, that copy testing can undermine the ability to produce breakthrough creative ideas."
He said that results of a recent study conducted by Unilever in cooperation with testing service Ameritest and examining three different testing systems found little relation between next-day recall and "the emotion in advertising that builds brands."
Indeed, the study found a negative correlation between how well consumers recalled ads and how well they liked them. Among the findings: Consumers' visual memories of sequences in ads they like is far stronger and lasts far longer than their ability to verbally recall ads they've seen.
The Four A's subcommittee on copy testing hopes to rectify the situation by "developing a new tool that would yield a measure for emotional responses to advertising," said Michael Donahue, exec VP-member services, either by funding a study or collaborating with a research organization.
Joe Plummer, exec VP-director of research and insight development, Interpublic Group of Cos.' McCann-Erickson WorldGroup, and a subcommittee member, said, "There's been no fundamental advance in copy testing since the early 1980s. So much new information is available today on how the mind works and processes emotional responses to stories, metaphors and symbols."
Copy testing has huge implications, maintains Robert Deutsch, a Boston-based researcher and consultant on neuroscience and anthropology who has studied how people attach themselves to brands. He believes that recognition and recall testing do not measure the things that produce attachment. "Brand is that relationship with a thing, such that your emotional identification with it, such that no matter what, you cannot be without," said Mr. Deutsch. "Brand is a way of saying, `I don't care what happens, I cannot live without you.' "
Like Unilever, Procter & Gamble Co. is investigating brand "equities in a more emotional way," said Jim Stengel, global marketing officer at the consumer-products giant. "We have a proprietary tool where on a periodic basis we are measuring equities, and much more than does it clean better and make this garment whiter, but much more about, am I connecting with it as a consumer? I'm happy with where our equity measurement is going. We need to make sure we're on a parallel path with copy measurement."
David Brandt, exec VP-global clients at researcher Ipsos-ASI, said "recall testing never was intended to measure emotions. By the same token, emotional response variables don't measure the ability of advertising to leave an impression in the marketplace. So there's no one approach. You need both measures."
But he added that big global clients are demanding better ways to measure emotional impact. "You'll be hearing a lot about this over the next 12 to 18 months as things evolve in that direction."