Using Twitter to model consumer perception and product development opportunities: A use case with Turmeric

Abstract

Turmeric has been promoted as a functional ingredient due to its many suggested health benefits, such as alleviating inflammation, pain and digestive disorders. As with most functional ingredients, it can be difficult to fully understand how consumers use and perceive these ingredients. Sensory Scientists are often tasked with the job of summarizing these consumer perceptions. However, extensive resources are needed to execute large consumer tests, inevitably limiting the scope of the study. Alternatively, assessment of consumer commentary about key ingredients on social media can be utilized as a valuable and efficient tool to overcome consumer testing challenges. Using the R package, rtweet, a total of 20,000 Tweets were collected that contained the keyword “Turmeric.” A machine learning technique known as topic modeling was used to illuminate abstract themes from the collective Twitter commentary. Consumer perception including utility and lifestyle, as well as, white spaces for product development were discovered. With this cost-effective and simplistic method, Sensory Professionals can be better positioned to understand consumer perception of ingredients on a greater and more holistic level.