The Research of Portal Competition Based on Game Theory

Ye Wang, Ting Wu, Yaqian Zhu, Yu Zhang, Yang Liu2016 · DOI

Abstract: Based on the theory of competition, social network analysis, and game theory, this research constructs a game model for portal competition analysis. By studying the current status of portal competition from a new perspective, the paper introduces a novel analytical tool for the industry, opening up new ideas for competitive innovation in a crowded market.

Introduction

The web portal industry is currently grappling with a "homogeneity" crisis. Most major platforms offer nearly identical content, leading to stagnant audience growth and aggressive price wars for user attention. Traditional market analysis often treats portals as monolithic entities, failing to capture the nuances of how users switch between specific content channels, such as News, Finance, or Sports, across different platforms. This study aims to provide a more granular toolkit for portal enterprises to understand their actual competitive positioning and identify opportunities for strategic differentiation.

Methodology

The researchers employed a multi-stage hybrid methodology combining Social Network Analysis (SNA) with the Cournot game theory model:

  1. Empirical Survey: Data was collected from 190 questionnaires regarding user preferences across eight specific channels (News, Video, Financial, Entertainment, Sports, Auto, Real Estate, and Science/Tech) for ten major portals, including Tencent, Sina, Netease, and Sohu.
  2. Social Network Analysis (SNA): Using Ucinet software, the survey data was transformed into relationship matrices to visualize the "competitive network." This step identified which portals are intimately connected (frequently chosen together) and which are isolated.
  3. Clustering: To manage the complexity of the game theory calculations, portals were grouped into clusters for each channel based on their competitive proximity.
  4. Cournot Model: The authors applied the Cournot competition model—traditionally used for firms competing on output quantity—to these portal clusters. "Input" was defined as measures taken to improve competitiveness (e.g., timeliness of information). By setting marginal costs and revenue functions, the researchers calculated the Nash equilibrium, the point where no player can increase their yield by changing their strategy unilaterally.
Table 1: Clustering of portals across different content channels to simplify game theory modeling.

Mapping the Competitive Landscape

The SNA revealed a "Keen Competition" core within the Chinese portal market. Tencent and Sina emerged as the most "intimately connected" entities, indicating a high state of direct competition. In contrast, portals like 21CN and TOM were found to be in a state of relative isolation, suggesting they pose a minimal threat to the market leaders in their current configurations.

Figure 1: Social Network Analysis (SNA) visualization showing the competitive connections and proximity between major Chinese portals.

Key Insights and Game Theory Results

The application of the Cournot model yielded several critical insights into the profitability and strategic value of different content channels:

  • The Power of Integration: The "Integrated Weighted Channel" (a composite of all content types) produced the highest overall yield (4726). This suggests that despite the rise of niche sites, users still heavily value a comprehensive, "one-stop-shop" experience.
  • Channel Dominance: The study highlights that News, Video, and Entertainment are the primary drivers of user concern. Portals that optimize these specific channels, such as Phoenix and Sohu, see significantly higher theoretical earnings in the game theory model.
  • Earnings Leaders: In the comprehensive model, Sohu achieved the largest total earnings (9862), followed closely by Phoenix (9798) and Sina (9578). TOM recorded the lowest yield at 6024.
  • Nash Equilibrium in Action: The authors successfully calculated specific strategy combinations for each channel. For the News channel, the equilibrium strategy combination resulted in an overall yield of 4360, providing a benchmark for optimal resource allocation.

Conclusion

This research demonstrates that competition in the portal industry is rarely a simple "one vs. one" battle; rather, it involves dynamic clusters of competitors. While the study is limited by a relatively small sample size (190 surveys) and simplified marginal cost assumptions, it provides a robust framework for understanding market dynamics. Future work should expand this model to include modern social media platforms and short-video apps, which now compete for the same user attention that traditional portals once dominated.

Summary updated

Abstract

Based on the theory of competition, social network analysis and the game theory, this thesis builds up the game model of portal competition analysis, studies the current status of portal competition from a new viewing angle, to add a new analytical tool and viewing angle for the research of portal competition and open the idea of competitive innovation for the entire portal industry.
BibTeX
@misc{wang2016research,
  author = {Ye Wang and Ting Wu and Yaqian Zhu and Yu Zhang and Yang Liu},
  title = {{The Research of Portal Competition Based on Game Theory}},
  year = {2016},
  doi = {10.2991/icca-16.2016.93},
}
Figure