The Press, November 2008

If Barack Obama is successful in tomorrows’ (New Zealand time) presidential election, it will be in spite of something called the Bradley effect. The nature of this phenomenon offers some insights into the human mind that are relevant to any marketer.

It is named after African-American politician Tom Bradley who failed in his bid to become the Governor of California in 1982. This was despite holding a significant lead in the opinion polls up to election day. What later research revealed was that many white voters didn’t want to appear prejudiced when questioned by pollsters so indicated support for Bradley, but in the privacy of polling booth went for the white candidate.

What this shows is that although we like to present ourselves as smart, rational and balanced, the way we behave can be a lot less logical.

Our behaviour can also be manipulated. A study cited in Malcolm Gladwell’s book Blink had a group of African American students sitting a standardised test. A sub-group were required to write their race on the test form, and recorded significantly worse marks than the rest of the sample. Researchers found that the negative stereotypes associated with African American educational achievement had a subconscious effect on their performance.

What’s all this got to do with marketing?

A lot, if a new book by Martin Lindstrom called “Buyology: Truth and Lies About Why We Buy” is to be believed. Based on a three-year, $7 million study, it has some fascinating insights into the subconscious thinking behind our purchase decisions.

For example, Lindstrom’s studies revealed that the graphic health warnings on cigarette packets may actually encourage smoking! They found smokers’ brains quickly edited out the gross pictures and consequently each time they saw the image it actually made them feel like smoking. It became an abstract signal to their brain to light up.

To combat this effect Lindstrom is helping design cigarette packets that change regularly. Change in colour, appearance and warnings so the smokers’ brains don’t get used to it.

The lesson for marketers is what people think is less important than what they actually do.

Research focused too much on customers’ rational response to a product has led to some of commerce’s most spectacular disasters. The infamous New Coke debacle, Millers short-lived experiment with clear beer and Apples failed hand-held computer the Newton are just a few.

So why bother with researching your market at all?

Market research should never be used as a substitute for decision-making. The purpose of good research is to educate your decision-making about a marketing question, not replace the balance of experience, intuition and judgement a business owner must use.

It is not that research can’t be useful or valid; it is the way it is conducted that makes the difference.

Completed properly, research can provide real insights. Into a customer’s problems and needs; into what value they ascribe to our product and therefore how we can price it; about what other choices to solve their problem other than your product; into what process they use to evaluate and buy your kind of product.

It is easy, and risky, to build a marketing programme on too many assumptions. Good research will typically confirm many of these assumptions, but will also highlight a few things you might not have anticipated and in doing so make a real difference. Fundamentally research can be of value because you don’t know what you don’t know until you ask.

The lesson from the Bradley effect is that research often gets derailed when you ask customers a direct question. For example you might get a great result in surveys asking whether they would like to buy the widget you are making, but poor returns when you go to sell. If instead you try and really understand the needs they have and use your knowledge and skills to provide a solution to those articulated problems, you are much more likely to get a positive result.

So how do undertake an effective research process? The best place is to start at the end:

  • What decisions do you want to make: how will you actually use the information you collect – to enter a new market, introduce a new product, change a pricing strategy. The clearer you are on the outcome the better result you are likely to get.
  • What information is required to make these decisions: is it understanding the customer’s buying process, or the replacement cycle of your product with customers.
  • How can you best get that information: from industry sources, direct interviews, online surveys and so on. Different methods gather different types of information.
  • Who do you need to talk to: what’s a representative sample of your market, what mix of job titles do you want, who do you need to exclude?

Political polls seem to be poor predictors of election outcomes. That’s because we are not necessarily the rational, logical people these polls assume us to be. Research can make a real difference to marketing but only if done carefully, for the right reasons and in the right way.

Owen Scott is from marketing company Concentrate Limited. He’s an undecided voter.

Subscribe to our blog