AI-Driven Predictive Modeling for Customer Acquisition in Digital Payment Platforms

Abstract

This study examines the impact of predictive modeling on customer acquisition rates in digital payment platforms, addressing the practical problem that many fintech firms generate large volumes of transaction and behavioral data but still struggle to convert those data into efficient customer acquisition strategies. The purpose of the research is to determine whether predictive modeling, supported by machine learning based transaction analysis, improves customer targeting efficiency and acquisition outcomes in cloud-based and enterprise digital payment cases. Using a quantitative, cross-sectional, case-based design, the study collected data from 218 respondents drawn from cloud and enterprise payment platform contexts, including professionals in data analytics, marketing, operations, product, and management roles. The core variables were predictive modeling, machine learning based transaction analysis, customer targeting efficiency, and customer acquisition rate. Data were analyzed using descriptive statistics, Cronbach’s alpha, Pearson correlation, and regression analysis. The findings show strong reliability across constructs, with Cronbach’s alpha values ranging from 0.838 to 0.881, and an overall scale reliability of 0.889. Descriptive results indicated high agreement across all variables, including predictive modeling (M = 4.18, SD = 0.61), machine learning based transaction analysis (M = 4.11, SD = 0.66), customer targeting efficiency (M = 4.07, SD = 0.64), and customer acquisition rate (M = 4.14, SD = 0.59). Correlation analysis revealed significant positive relationships between predictive modeling and customer acquisition rate (r = 0.681, p < .001), transaction analysis and customer acquisition rate (r = 0.643, p < .001), and targeting efficiency and customer acquisition rate (r = 0.701, p < .001). Regression results further showed that predictive modeling alone explained 46.4% of the variance in acquisition outcomes, while the combined model of predictive modeling, transaction analysis, and targeting efficiency explained 59.3% of the variance, with targeting efficiency emerging as the strongest predictor (β = 0.374, p < .001). The study implies that digital payment firms can improve acquisition performance by embedding predictive analytics into evidence-based targeting and campaign decision systems.