50 years using the wrong model of TV advertising

Robert Heath and Paul Feldwick

This article summarises a paper to be presented at the 50th Market Research Society Conference. It investigates the dominance of the information-processing model, which has been assumed to be correct since the first UK TV ad ran in 1955. We present evidence that the model is over-simplistic and contrary to learning from psychology, and examine why it has never been successfully challenged. Finally we present an outline for a new model, and summarise some implications.


In 1999 a launch TV commercial for a snack food product aimed at teenagers was pre-tested. It consisted of a song with meaningless gibberish lyrics, accompanying a series of surreal scenes. It was pre-tested amongst teenagers using familiar 'impact and communication' questions, and average scores were produced for constructs including 'ease of understanding', 'believability', 'relevance', 'branding' and 'persuasion'. Not surprisingly, scores were below norms.