Dynamic pricing of new experience products with dual-channel social learning and online review manipulations

Online reviews normally come from two distinct sources: first-party reviews are reviews published (by consumers) on a firm's own platform, and third-party reviews are ratings and feedback generated on a third-party website (e.g., a social media profile). Manipulations of first-party reviews could affect their credibility. We consider a two-period monopoly dynamic pricing problem with dual-channel social learning (SL) and truncated review manipulations (i.e., a firm may delete extremely low and high ratings). We propose a critical measure for SL outcome (SLO) that gauges consumers’ quality evaluation through SL and drives equilibrium outcomes. We first consider the case of a firm with myopic consumers. Without manipulations, we find that the optimal policy typically consists of increasing and decreasing prices regarding a threshold structure of SLO. The optimal price, expected profit and consumer surplus are monotone in SLO. More accurate dual-channel reviews benefit the firm and its consumers. With manipulations, we characterize the optimal price path in closed form using a novel index of manipulated SLO. Manipulations yield a higher expected profit increasing with manipulated SLO, but can induce three outcomes: benefiting all consumers, or benefiting some but hurting the others, or hurting all. More first-party reviews always facilitate the firm, but can benefit consumers only under weak manipulations. However, more third-party reviews can be detrimental for the firm but conducive for consumers under strong manipulations. We also discuss the robustness of our main qualitative insights and some extensions with additional salient features. Research implications and future directions are discussed finally.