Technology is opening up all sorts of new avenues for market researchers, but ultimately they’re still relying on the quality of the data they collect and how they interpret that to deliver useful insights to clients. But the data is often, of necessity, gathered in an artificial environment and can that be used to predict what real consumers will do?

“It’s critical that the consumer data we get is actionable,” Fabien Craignou, director consumer insights at Firmenich, told the IIeX conference (Amsterdam, February 2019). Firmenich creates fragrances for anything from perfumes to shower gels to laundry detergents. “We want fragrances that are preferred by consumers and that are also able to support product benefits or brand values like elegance, sophistication, cleanness,” he explained.

When it comes to research, he’s demanding. “We want to have consumers who are engaged on the task, who are focused – and discrimination is expected.” By which he meant that it’s necessary to extract more than binary like/don’t like responses or basic descriptive profiles of the scents being tested. “We want to see differences in order to select the perfumes which have the best potential for the market.”