Platform’s Recommendation Strategy Considering Limited Consumer Awareness and Market Encroachment

Abstract

In today’s increasingly interconnected world, the platform owning the private label product (PL product) has formed a cocompetition relationship with the manufacturer, which may affect its recommendation strategy. Considering consumers’ limited awareness and relative preference, a game model is established to study the impacts of the platform’s recommendation strategy on pricing, consumer surplus, and social welfare. We demonstrate that consumer preference does not absolutely determine the platform’s recommendation strategy. When consumers prefer the manufacturer’s product (MB product), the platform recommends the PL product if the commission rate is low; otherwise, it recommends the MB product. When consumers prefer the PL product, the platform is willing to recommend the MB product only if the PL product’s valuation is not too large and the commission rate is high. By contrast, the manufacturer always prefers the MB product to be recommended. Furthermore, the recommendation strategy selection has different effects on different types of consumers. Although uninformed consumers always prefer the high-value product to be recommended, informed consumers may prefer the platform to recommend the low-value product. More importantly, the platform can achieve Pareto improvement by providing paid recommendation service, benefiting itself and the manufacturer. We also extend our model to conduct further analyses (e.g., differentiated consumer awareness levels, positive production cost, and corporate social responsibility), which examines the robustness of our results. Our study helps to explain the market practices and provides valuable guidelines for platform encroachment and recommendation decisions.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 18 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.