Article: Friday, 8 June 2018
Brand managers need to know how consumers perceive their brands. Do consumers actually have the ‘right’ associations with brands as intended by the companies? And are these associations consistent, or are they vastly different across consumers? Until recently, all that brand managers could do to find out, was to trust what consumers told them. But no longer. Researchers Hang-Yee Chan, Maarten Boksem and Ale Smidts from Rotterdam School of Management, Erasmus University (RSM), literally looked inside people’s heads and discovered that brand image and brand image strength are clearly visible in our brains.
Marketers try to instil specific mental associations about their brands in consumer’s minds. A beer brand for example could try to evoke images of trendy people at a party, and a cereal brand of a loving family at the breakfast table. But how effectively and consistently these images are forged in consumer’s minds has always been hard to measure. The best marketing researchers could do, was work with self-report instruments, such as surveys. However, brand image surveys require people to verbalise what is essentially an implicit and visual experience (the ‘picture’ the brand generates in a consumer’s mind), which makes such measures not perfectly reliable.
Hang-Yee Chan wanted to know if there was a more reliable way to find out which associations people have with brands. He realised this could potentially be done by directly looking for these associations in people’s brains. Thereby circumventing the need for self-report. So, he set up an experiment, where he put people in a functional magnetic resonance imaging (fMRI) scanner to have a look. By applying machine-learning techniques, he was able to extract the ‘neural profile’ of brands. That is, how the images people construct in their minds, when they think of a certain brand, relate to other types of imagery, like scenes of parties, family scenes, work scenes, etc.
Chan asked 38 participants to perform a visualisation exercise involving 14 well-known brands, such as Apple, Disney, Heineken, Kellogg’s, Microsoft and Durex. The participants were asked to construct a mental picture that best fit the brand’s intended imagery and message. Next, they were put in an fMRI scanner and had to recall the mental images they had constructed, while they saw the 14 brand logos. Then, they viewed 112 pictures depicting four different social settings: people working in an office, partying with friends, being intimate with a romantic partner, and being with family.
Hang-Yee Chan then analysed to what extent activity in the participant’s brains, while watching the brand logos, resembled brain activity while viewing pictures of the social settings. That way, he created a neural profile for each brand, for each participant. Essentially, this means that a brand acquires a brain ‘similarity score’ on each of the four social settings. This score then indicates to what extent the association with the brand (the ‘brand image’) resembles the imagery depicted in the social scenes. For example, this revealed that the brand image of Heineken was closely associated with ‘Party’, the image of Kellogg’s resembled ‘Family’ and Microsoft was strongly associated with ‘Work’. Notably, these associations were acquired without ever asking participants directly what their associations with those brands were. This way, problems with potentially biased self-reporting, were circumvented.
Chan then looked at how alike these brand associations in the brain were across individuals. He discovered that brands that evoke very similar associations in the brain across individuals, such as Heineken, Durex and Disney, have a much stronger brand image than brands that showed very diverse brain responses.
Chan then used the neural data to predict co-branding suitability. If the neural profile of two brands is very similar, then these two brands can be marketed together effectively, like Apple and the headphones brand Beats. Brands that evoke very different brain responses are much less suitable for co-branding, such as Kellogg’s and Durex.
This novel method is very flexible. In this study, four social categories were chosen to measure brand image and to classify brands, but marketers are free to choose whatever categories they consider relevant for their market. One just needs to find stimuli, such as pictures, words, emoticons, sounds, or even smells, that are thought to represent their brand image associations as best as possible, and then test how closely these stimuli resemble their brands in the brain.
The time when marketeers could only rely on what consumers told them about brands seems to be ending. Hang-Yee Chan: “By demonstrating a novel approach to capture consumers’ visual representations of brand image, this study represents a first step towards understanding sensorial consumer knowledge and experience.”.
Rotterdam School of Management (RSM)
Rotterdam School of Management (RSM)
King's College London
RSM offers Executive Education and Master programmes in various business areas for any stage of your career. For instance:
Corporate Communications & PR Manager
Science Communication and Media Officer