Articles

A Practical Guide to Collections Stages in Lending

Collections works best when it matches the delinquency stage, the product, and the customer profile. One contact strategy for every case creates noise, lowers customer experience, and wastes recovery effort.

This article breaks down pre-collections, early collections, late collections, and refinancing, with a focus on decision logic, segmentation, and disciplined communication.

Antifraud investigation in lending: how to detect, trace, and validate risk

How to Monitor Scorecards and Catch the Signals That Matter

A scorecard can look healthy on a dashboard and still drift in ways that change decisions. The key is to monitor more than one metric: shifts, attribute stability, bin stability, predictive power, rank order, and the behavior around cutoffs.

Portfolio management in lending: risk, funding and profit

Learn how to read risk measures alongside customer economics and funding costs. Use them to decide which cohorts and products earn their place in the book.

A Practical Map of Credit Risk: What Actually Affects Loan Performance

Credit risk is not just the chance that a borrower will miss a payment. In practice, loan performance depends on a wider system of risks that shape origination quality, collections, funding stability, compliance, and day-to-day operations. This article maps the main risk categories around credit risk and shows how they interact in lending businesses.

How to implement an automated decision strategy that keeps working under failure

Automated decision strategy is not just about getting the happy path right. It also needs to keep making safe, explainable decisions when data is missing, providers time out, or a complex decision flow fails. This article shows how to design decision logic with failover and missing data strategies built in from the start.

Attribute and Model Management: How to Track Stability and Preserve Your Decision Strategy

Adding more attributes to a model does not always improve decision quality. If predictors are poorly grouped, weakly represented, or unstable over time, they can degrade model performance and create risk for the wider decision strategy. This article explains how to manage attributes and models with stability in mind, and why binning and categorizing predictors remains a practical way to keep automated decisions explainable, traceable, and reliable.

Cure rates fell in bucket 2. Which release booked those loans?

A portfolio number moves and eight things could have caused it. Only one resolves to a yes or a no from records you already hold: whether the decision logic that approved those loans differs from the logic running now. This is how to check, what a release snapshot contains, and where the record stops.

What should an anti-fraud team own and what should it buy?

Anti-fraud is not a product you install once and forget. It is an operating capability built from data, decision logic, traceability, and continuous tuning across the customer journey. Companies can buy parts of that stack, but if they outsource the core logic and enriched data, they often create a lock-in problem that is hard to unwind later.

The True Nature of Fraud and How to Build Anti-Fraud That Works

Most anti-fraud programs focus on single bad applications, stolen identities, or suspicious transactions. That matters, but it misses the main point. The real fraud problem is not isolated opportunists. It is organized actors who treat fraud like a business, calculate return on investment, and target the weakest operators first.