Social media analysis offers businesses a new way to forecast feature demand

A recent study from the University of Florida's Warrington College of Business reveals that social media posts can provide insights into consumer demand for specific product features. The research, published in Production and Operations Management, demonstrates that social media content evoking consumers' purchase intentions can be linked to higher product sales.

The study utilized an AI deep learning framework to analyze over 3 million Instagram posts and sales data from a kitchenware manufacturer. By creating a measure called "purchase-evoking frequency" (PEF), the researchers identified posts that could signal consumers' intent to buy. Products featuring a particular attribute, such as color, that were frequently mentioned in purchase-evoking posts had higher sales.

The relationship between PEF and sales varied depending on the type of product. For "search goods," which have features that can be evaluated before purchase, the effect of color PEF was stronger. This contrasts with "experience goods," where customers place a higher value on functionality and intrinsic quality. Additionally, the impact of color PEF was more pronounced for online sales compared to offline sales.

The researchers suggest that businesses can leverage these feature-level signals from social media to guide product design decisions, inventory planning, and supply chain coordination. By focusing on purchase-evoking posts rather than relying solely on broad measures of social media engagement like likes or shares, companies can better understand customer preferences and improve their operations.