Product line design under preference uncertainty using aggregate consumer data

This research studies the product line design problem when consumers are subject to perceptual errors in assessing their intrinsic preferences. If perceptual errors are driven by common variables, then a firm may use aggregate consumer data (e.g., conjoint studies or anonymous usage data) to deduce the errors and infer the consumer preferences. In this way, we develop microfoundations necessary to show when and how the firm can understand consumer preferences better than consumers themselves, a situation we call superior knowledge. But is superior knowledge ever unprofitable? How should the firm with superior knowledge design its product line? Do consumers receive more-relevant products or simply have more surplus extracted? Can data collection help consumers make better choices? Our results suggest that consumers’ rational suspicions may prevent the firm from exploiting its superior knowledge. In addition, the burden of signaling may force the firm to offer efficient quality for its products. Therefore, allowing the firm to collect aggregate consumer data may be strictly Pareto improving.