having said it are only fair’ 16 Research can tell you which of several television commercials will sell the most. What is the best technique for pre-testing television commercials? This is the most controversial issue in the advertising business, but there is common agreement among researchers that testing for recall is for the birds. Yet, for reasons which escape me, most advertisers still insist on using it. It has four shortcomings: A Nobody has been able to demonstrate a relationship between recall and sales. B Some commercials which score about average on recall, score below average on their ability to change the viewer’s brand preference. Celebrity commercials, for example, usually score above average on recall and below average on changing brand preference. C It is too easy for the copywriter to cheat. ‘When I want a high recall score,’ says my partner David Scott, ‘all I have to do is to show a gorilla in a jock strap.’ D It is open to question whether recall tests even measure recall. I believe they measure the viewer’s ability to articulate what he or she recalls, which is a very different thing. For all these reasons, I prefer testing methods which measure your commercial’s ability to change brand preferences. Research can measure the wear-out of your advertising. For five years, the theme of Shell’s commercials was mileage, and tracking studies recorded increasingly favorable attitudes to the product. When attitudes finally stopped improving, the advertising was changed from demonstrations of mileage to consumer testimonials, and the upward trend was resumed. 17 Research can tell you how many people read your advertisements, and how many remember them.