Speaking at our Advertising Research conference in London today, McEntaggart used the tool analogy to illustrate the different uses for pre-testing research.

“On one hand there’s the lump hammer,” he said, “which is very heavy handed and destroys, while on the other is the masonry or carpentry hammer that you use to build things.”

McEntaggart says the latter should be the tool of choice when testing new adverts. In that way, he said, the research is part of the creative process, part of the development process, and definitely not “just a box-ticking exercise”.

And pre-testing “should be done at an early stage”, he said. To explain why, McEntaggart recounted the history of two adverts for one of PepsiCo’s flagship brands – Walkers’ crisps.

Walkers launched a new range of Baked crisps in January 2007 accompanied by a TV campaign that showed the gloomy and rainy UK being dragged over the Atlantic to be joined with Jamaica. The firm had earmarked female dieters between the ages of 25 and 44 as the target market, and the campaign launched at the time of year when people are trying to stick to a more healthy lifestyle after the excesses of Christmas.

However, the firm soon found that the biggest buyers of the new crisps were men aged between 25 and 44 who professed that “looking good” was one of their main priorities in life.

So PepsiCo quant-tested a new ad in 2009, which aimed to appeal more to this market, but the testing found that the music was dated, the final scene was not funny enough and, crucially, the message about the crisps’ low fat content was not prominent enough.

Taking the findings into account, a new advert was created with a more up to date soundtrack and a funnier scene – brand ambassador Gary Lineker wearing a muscle suit that deflates on a bus – while the message about the crisps’ fat content was made clearer.

The results: 144% more ROI than the original ‘Island’ campaign, said McEntaggart.