Determinan Kredit Macet dan Perencanaan Desain Sistem Analisis Kredit Fintech (STUDI KASUS PT XYZ)

Abstract

Kredit macet merupakan salah satu risiko utama pada perusahaan fintech pembiayaan yang dapat memengaruhi kualitas aset dan keberlanjutan operasional perusahaan. Penelitian ini bertujuan menganalisis pengaruh nilai pembiayaan, status pembayaran, dan jenis produk pembiayaan terhadap kredit macet serta merancang sistem analisis kredit berbasis Early Warning System (EWS) pada PT XYZ. Penelitian menggunakan pendekatan kuantitatif dengan analisis Binary Probit terhadap 1.271 data pembiayaan tahun 2025. Hasil penelitian menunjukkan bahwa nilai pembiayaan, status pembayaran, dan jenis produk pembiayaan berpengaruh signifikan terhadap probabilitas terjadinya kredit macet. Berdasarkan hasil tersebut, penelitian dilanjutkan menggunakan metode ADDIE yang dibatasi pada tahap Analysis dan Design untuk merancang Credit Risk Assessment Engine. Sistem yang diusulkan mengintegrasikan proses validasi data, penilaian indikator risiko, perhitungan Eligibility Score, pembentukan Risk Profile, penyajian Early Warning berbasis indikator warna, serta rekomendasi pembiayaan sebagai pendukung keputusan Credit Risk Officer dan Supervisor. Rancangan sistem diharapkan membantu perusahaan melakukan penilaian risiko secara lebih objektif, konsisten, dan terstandarisasi.

Loan default is one of the major risks faced by fintech financing companies, as it may affect asset quality and business sustainability. This study aims to examine the effects of financing amount, payment status, and financing product type on loan default and to design an Early Warning System (EWS)-based credit analysis system for PT XYZ. A quantitative approach was employed using the Binary Probit model with 1,271 financing records from 2025. The results indicate that financing amount, payment status, and financing product type significantly influence the probability of loan default. Based on these findings, the study continued using the ADDIE method, limited to the Analysis and Design stages, to develop a Credit Risk Assessment Engine. The proposed system integrates data validation, risk indicator assessment, Eligibility Score calculation, Risk Profile classification, color-based Early Warning, and financing recommendations to support decision-making by Credit Risk Officers and Supervisors. The proposed system is expected to assist the company in conducting more objective, consistent, and standardized credit risk assessments.