Most researchers have at one time or another been unsure about whether a consumer's thoughts on a product were genuine or overstated. It’s easier to dig a bit deeper in qualitative research, but in the quantitative environment researchers still find themselves having to take what people say at face value.
“Maybe if we try to analyse how people communicate about their attitudes, their needs and their interest, we might be able to uncover the underlying emotions and predict interest with greater validity,” suggested Samantha Bond, senior researcher and London team lead at Skim. But how can one go about doing this?
At MRS Impact (London, March 2019), voice was offered as one possible route. Skim partnered with a German startup audEERING, which uses machine learning algorithms to detect the emotions from voice, based on subconscious reactions. It’s already being used in call centres to detect when a consumer is unhappy, and allow a more senior colleague to step in and help out if necessary.
“We wanted to take it one step further, and see how we could apply this really great new innovation in market research,” explained Judith Suttrup, research manager at Skim. All that was needed was a client – in this case Johnson & Johnson.