Enhancing Crowdfunding Success With Machine Learning and Visual Analytics: Insights From Chinese Platforms

Crowdfunding helps SMEs and entrepreneurs secure financing, but many projects fail. This study examines how technological features of crowdfunding platforms, specifically advanced analytics and visual tools, influence project success. Using data from 2,285 projects (2022–2024) on Chinese platforms Xiaomi Youpin, Modian, and Zaodian Xinhuo, we apply a Random Forest model to identify key predictors of success. We employ the Investment Attractiveness Matrix to visualize relationships among project innovativeness, risk, and success rates. Our findings indicate that technological features significantly enhance success rates. The Random Forest model highlights critical factors such as project quality signals and social capital. The Investment Attractiveness Matrix aids creators in strategically positioning their campaigns. This research contributes by integrating machine learning and visual analytics into crowdfunding studies, offering a novel approach. It extends the Resource-Based View by demonstrating that technological resources are critical capabilities for platforms. Applying Signaling and Social Capital theories, we show how presentation and networks impact success. Practically, the study provides insights for platforms and project initiators to optimize strategies through advanced analytics.