Blueprint for a Priming Study to Identify Customer Needs in Social Media Reviews

摘要

Unbiased customer reviews in social networks may hold the key for innovations in the saturated market of consumer goods. Customer reviews do not only offer information directly about the product, they also provide insights into the user's environment, customer habits and usage behaviour. These latent needs are stated objectively in reviews. This study aims to overcome the weakness of state-of-the-art machine-learning algorithms that can only extract explicitly stated needs. Key part of the study is the developed method to record and evaluate the reaction time of test subjects to analyse the association between a latent need category and a related word. As a result, we obtain word clusters that express an association with a latent need and a blueprint for upcoming studies that focus on the extraction and utilizing these needs. This knowledge can be used for further research in an automated need identification process and customer driven production.