鉴于交叉网络效应导致用户加入双边平台的效用随用户规模动态变化，提出根据用户规模进行适应性动态定价的策略思想，并运用数值计算方法对该定价策略的效果进行深入研究。首先，引入平台动态竞争建模方法，构建了包含用户规模的双边平台适应性动态定价模型；接着，根据数值计算结果对动态定价与静态定价的效果进行比较；最后，考察了平台竞争主要参数的变化对动态定价策略效果的影响。研究表明：（1）动态定价显著优于静态定价，模型主要参数的取值变动不会改变动态定价具有相对优势这个定性结论；（2）提升服务质量或改变基准用户数不会明显增加动态定价的相对优势，但强交叉网络效应或前瞻性用户都会增强动态定价的相对优势。研究结果有助于平台企业管理者更好地制定平台定价策略。

In managerial practice, platform companies often employ price subsiding strategy to attract users boarding on platforms, which can overcome utility deficiency problem caused by too small user base. Evidently, the subsiding strategy implicitly contains the rationality of price changing with user base. Literatures in two-sided market emphasized the essential characteristics of user utility increasing with user scale dynamically, but rare literatures had examined how to improve company profit by exploiting that characteristic. Considering that, the adaptive dynamic pricing strategy based on user scale is put forward, and the effect of pricing strategy is studied by numerical calculation. Firstly, platform pricing is assumed to be the function of user scale, and a dynamic competition model coupled user scale and platform price is developed to analyze the competitive dynamics of two-sided market. The model assumes that there are two competitive platforms existed in the market, i.e., the new entry platform E and the old platform I, and the quality of the former platform is superior to the latter one,conversely the initial user scale is. The two platforms compete directly for the same user community. Secondly, based on the model, the dynamic pricing strategy is analyzed from three aspects: (1) solving the optimized dynamic price and pricing coefficient, and uncover that the optimized solution depends on specific parameter values based on the optimized dynamic pricing; (2) combining every possible couples of parameters from three of relative quality advantage, cross network effects, and user prospective coefficient, and comparing the optimized solutions of dynamic pricing strategy and static pricing strategy by varying different parameter combination value so as to demonstrate the reliability of the relative advantage of dynamic pricing strategy; (3) examining the influences of the main competition parameters (relative quality advantage, cross network effects, user prospective coefficient, and initial user scale) on dynamic pricing, and comparing the effects and influences of different parameters on dynamic and static pricing, further demonstrating the superiority of dynamic pricing than static pricing. Research results show that: (1) dynamic pricing is significantly better than static pricing strategy, which will not be affected after changing the model parameter; (2) service quality improvement or changes of benchmark users do not increase the comparative advantage of dynamic pricing strategy, but stronger cross network effects or more prospective users are benefit to the comparative advantage of dynamic pricing strategy. The results of this study can help managers develop better dynamic pricing strategy.