Thursday, March 11, 2010

Importance of experimentation to behavioural economics

Prompted by such books as 'Predictably Irrational', the application of behavioural economics is quite in vogue as it applies to advertising, marketing and market research. As Dan Ariely writes on page xii of this tremendous book:
"my goal is to fundamentally rethink what makes you and the people around you tick."
This theme has been picked up by Rory Sutherland (vice-chairman of Ogilvy's and president of the IPA) in his article 'Why advertising needs behavioural economics'; now also available on the site of the American Association of Advertising Agencies.

Robert Bain covered this subject in the February 2010 edition of Research magazine in a piece called 'Thoughts into Action'. What I found curious about this article was its comment about the methodological implications of behavioural economics. He wrote that:
"There are signs, however, that the MR industry is shifting its focus towards qualitative techniques such as ethnography, which both Fiona Wood and Les Binet pick out as one of the areas where behavioural science can be usefully applied. When Research asked some industry voices at the start of the year for their predictions for market research in 2010, the most prominent theme that came through was the rise of qual."
Without wishing to detract at all from what qualitative research has to offer, my interpretation is quite different. Following Dan Ariely's book, he makes quite clear, in his introduction, that his methodological approach is based on experimentation. He talks about the impact a Professor Frenk has had on his thinking when he would challenge Dan Ariely to devise an experiment to test any theory he had (page xv). On page xxi, he writes:
"For social scientists, experiments are like microscopes or stobe lights. The help us slow human behaviour to a frame-by-frame narration of events, isolate individual forces, and examine those forces carefully and in more detail. They let us test directly and unambiguously what makes us tick" 
And Dan Ariel's book is full of quantitative-style tests of people's reaction to various situations.

This is, for me, a key element of how research can be applied to a behavioural science approach. I can imagine that there could be frequent testing of all types of marketing and communications testing, based on specific experiments amongst the target population.  For very frequent testing, branded custom panels could be the way to go - as described recently by Vision Critical. Companies do not need to commission big quantitative projects to do this - ongoing quick experimentation can provide the appropriate answers.

And if you listen to Rory Sutherland's podcast interview with Robert Bain about behavioural economics, Rory Sutherland does talk about the need for solid, verifiable evidence which follows the science of behavioural economics. This is the type of evidence which will impress the boardroom - 'numbers' not 'mood boards'.

Methodological issues aside, I would totally concur with Rory Sutherland that behavioural economics potentially brings great value to the discipline of marketing. It helps demonstrate that marketing can be very important.