Beyond implicit associations, to implicit choice

Over the last two years implicit association tests have started to become a standard tool in market research. The 2017 GRIT report tells us 80% of researchers are using (53%), or considering (27%) nonconscious research methods, and that:

“Implicit/IAT is perceived to be one of the fastest growing nonconscious methods in the industry”

These methods are based on academic research originally done at Harvard University*, which measured unconscious racism or sexism by timing how fast people respond to different stimuli. Researchers found that the vast majority of people hold unconscious associations based on race, gender or age.

This opens a huge question: if people have prejudiced attitudes, does this mean they act in a racist, sexist or ageist way? The answer is not obvious – and in fact, more recent research summarised here has cast doubt on this. The correlation between racist attitudes and racist behaviour is lower than you might expect.

This led us to wonder about how implicit tools are being used for marketing and branding questions. These tools often uncover implicit brand associations (Coke is associated with “Authentic”, or BMW with “Exciting”). But we are now learning that these associations don’t necessarily drive behaviour. People might think Coke is authentic but not go ahead and buy it.

What you really need to know is not only what people think and feel, but how they will behave.

We realized that brands need a more direct way to predict what their customers will actually do. As a result, we have developed a set of tools for measuring implicit choice, rather than implicit association.

Implicit choice uses the same technique of measuring reaction time, to find out how intuitive people’s choices are, how confident they are in their judgements and how reliable those choices are. But the measures are choices – which can be used to directly predict market performance – rather than associations.

When linked with message priming, implicit choice allows us to test the impact of claims or concepts, and whether they work in changing customer behaviour (not just their attitudes or opinions, but what they will actually buy).

A popular method used by many of our clients is behavioural conjoint**, which combines conjoint analysis with implicit choice. This method tells us which product features or attributes most strongly drive consumer choice: does a price cut work better than a bigger pack or stronger claim? Or how much of a price premium can you earn by getting the messaging right?

No research method is perfect, and implicit choice is still only a proxy for what people choose in the real world. But this method gives much stronger correlations with true behaviour than either implicit associations, or traditional stated-preference survey research. Implicit choice measurement gives the best of both worlds.

Ultimately, choice is the only thing that matters for your bottom line – measure choice and you’ll be able to drive the business outcomes you want.

* Greenwald, A. G., McGhee, D. E., & Schwartz, J. L. (1998). Measuring individual differences in implicit cognition: the implicit association test. Journal of personality and social psychology, 74(6), 1464. ** Caldwell, L. (2015). Making conjoint behavioural. International Journal of Market Research, 57(3), 495-502.

Recent Posts

See All