Credit scoring is one of the most important technologies influencing the microfinance sector. This sector has grown rapidly and is regarded as a thriving industry. Credit scoring also represents both a primary source of revenue and a major source of risk for banking institutions. Classification With the desired accuracy of default risk yields significant economic benefits for banks. Conversely, misclassification of default can lead to revenue losses by rejecting creditworthy customers and to financial losses or non-repayment of granted loans by approving unsuitable borrowers. With the rapid growth of online lending platforms in recent years, the need for effective customer credit assessment has become one of the fundamental challenges in this industry, particularly in the areas of risk management and financial decision-making.

The present study adopts a meta-synthesis approach to analyze the dimensions of customer credit assessment in platform-based lending, with particular attention to customer Features and intelligent credit assessment approaches. Accordingly, the research methodology is applied in nature in terms of its objective and employs a meta-synthesis method for data collection. To select relevant studies, a systematic search was conducted in reputable databases (Scopus, IEEE Xplore, and ScienceDirect), resulting in the identification of 34 documents published between 2020 and 2025 as relevant and credible sources. These documents were subsequently reviewed and coded using the meta-synthesis method. The findings led to the identification of three main categories: (1) customer credit assessment objectives (2) intelligent credit assessment methods (3) customer features

The results indicate that hybrid approaches (quantitative methods such as machine learning combined with qualitative methods) and ensemble approaches (groups of machine learning and/or deep learning methods) achieve the highest accuracy in credit assessment. In addition, both hard and soft customer features such as account-related information and online behavioral data are utilized in credit scoring. Ultimately, a conceptual framework and model are developed based on the existing literature, in which the model is formed through the interrelationships among different credit assessment methods, customer characteristics, and the objectives of customer credit assessment systems.