Articles

Credit limit management for BNPL / Consumer lending

Credit limit management in BNPL and consumer lending is not a one-time check. It starts at underwriting, continues through upsell and cross-sell decisions, and should also run during portfolio monitoring as customer behavior changes. This article covers the main factors behind limit setting, when to increase limits, and when to reduce them.

Credit risk strategy in micro-lending

Micro-lenders face specific credit risks that demand a focused, operational strategy to cut default probability and improve repayment. This article defines four core risk groups (business, country, sector, borrower) and lays out practical controls: structured credit analysis, personal guarantees and collateral, LTV and term limits, risk-based pricing, portfolio reviews, provisioning, collections, and selective use of alternative data. It also shows how to encode these policies as auditable decision logic.

What data analytics areas need to be covered in the lending business - Decisimo

Lending decisions rely on more than credit checks. This article explains the data analytics areas lending teams need to cover, from marketing and sales to risk, fraud, portfolio management, and process improvement, and shows how each area supports better decision logic.

How a Rule Engine Helps Reduce Costs in Fintech

How to set up the process for credit approval

This article explains how to set up an 8-step credit approval process, from application to funding, and where each step should support automated, auditable decision logic. It covers verification, credit checks, underwriting, approval outcomes, documentation, and funding, and it notes that the steps vary by loan purpose.

Bridging Manual and Automated Scoring Models

Manual scoring models used statistical scorecards but slowed decision-making with time-consuming data work. This article explains how machine learning-based automated scoring speeds evaluations and improves accuracy, and why causal analysis and second-order effects matter when you implement decision logic and decision flows.

Creating Documentation for Credit Scoring Model

Clear documentation separates a usable credit scoring model from one that is difficult to defend. This article covers what to include, from management goals and sample selection to the sampling methods used in training, validation, and out-of-time testing.

Decision Engineer: role, skills, and impact in automated decision making

A decision engineer designs and builds decision models that automate decision logic. These models keep outcomes transparent, explainable, and aligned with business goals. This article covers the technical skills, including data analytics, programming, and machine learning, and the soft skills, including domain awareness, clear communication, and problem solving, needed to design and deploy reliable automated decision-making.

Champion-challenger tests with a 5% challenger can take 19.6 months

At a 4% bad rate, showing a one-point difference with a 5% challenger and a six-month outcome window can take 19.6 months. The split design also decides where applications with a missing split value go, and an open-ended challenger branch can send them to the challenger.

5 Rule Engine Architecture Components + Draft Testing

This article explains the five components of rule engine architecture: rule repositories, rule management, deployment interfaces, execution, and historical versioning. It also covers testing a draft decision flow before release using a saved test vector or supplied input, plus rule sets, batches, and case sandboxes.