Right-Party Contact Rate for US Bank Collections: From 26% to 47%+ with AI Propensity Timing
TL;DR What 26% Right-Party Contact Rate Costs a US Bank in Wasted Regulation F Frequency Budget Per Week A 26% right-party contact rate means that for every 100 outbound collection attempts, a bank reaches the actual account holder only 26 times. The remaining 74 attempts hit voicemail, wrong numbers, disconnected lines, or household members who […]
Collections Cost Per Recovery Benchmarks for US Banks: Pre and Post AI Deployment Data
TL;DR The gap between what US banks spend per recovered dollar and what that figure could be with governed AI in production is no longer theoretical. It is measurable, documented, and widening. Institutions still running manual dialler operations face a cost-per-recovery figure that has remained stubbornly anchored between $85 and $140 for more than a […]
MLOps for Indian Bank and NBFC Collections Models: Retraining Governance and RBI MRM Compliance
TL;DR A ₹7,200 crore NBFC deployed its AI collections model 14 months ago. The Gini has drifted from 0.74 to 0.61. No drift monitoring system was running. No retraining was triggered. No RBI MRM documentation was updated. The next examination is in 6 weeks. This scenario captures the core operational risk facing MLOps for Indian […]
Collections Cost Benchmarking for Indian NBFCs: AI vs Manual Agent Economics in 2025-2026
TL;DR The difference between a profitable collections operation and a loss-making one often comes down to a single number: cost per recovery. For Indian NBFCs managing portfolios of Rs. 500 crore or more, that number determines whether the collections function operates as a cost centre or a margin contributor. Collections cost benchmarking for Indian NBFCs […]
Collections Automation ROI for US Banks: 48% Cost Reduction with FDCPA Maintained
TL;DR Every collections leader at a US bank knows the number that keeps the CFO asking questions: cost per recovery. When that figure sits above $100, the collections operation is not just expensive; it is structurally misaligned with the margin profile of most consumer lending portfolios. The real return on collections automation for US banks […]
Right-Party Contact Rate for Indian NBFC Collections: AI Timing vs Manual Dialler Benchmarks
TL;DR Every collections operation has a single metric that determines whether agent hours convert into recovered rupees or vanish into unanswered calls and wrong-party conversations. For Indian NBFCs operating personal loan, two-wheeler, and microfinance portfolios, that metric is the right-party contact rate: the percentage of outbound attempts that reach the actual borrower. The gap between […]
Built From the Inside
What Seven Years of Waiting Taught Me About Building AI It was 2017 and I was in San Francisco to speak at a fraud analytics conference. I had a confirmed slot on the programme, I knew what I was going to say, and I had arrived well in advance. What I hadn’t anticipated was what […]
AI Charged-Off Debt Recovery for US Banks: Improving ROI on Legacy Portfolios
TL;DR A charged-off portfolio on a US bank’s balance sheet has a specific financial character. The accounts have been written down. The provision has been taken. The loss is already on the books. Recovery at this stage is pure upside against a cost base that has already been absorbed. The standard model for working that […]
AI Self-Learning Models for NBFC Collections: How the Technology Works
TL;DR A propensity model deployed on a personal loan portfolio in January was trained on 18 months of historical payment data. At deployment, its scores were accurate. Recovery rates improved through the first quarter. By August, the picture has changed. Payment rates in the mid-propensity band have dropped. The model continues to route accounts to […]
MLOps for SA Credit Providers: What AI Collections Implementation Looks Like
TL;DR Building an AI collections model and running one are two different disciplines. A model that performs well in development can fail in production for reasons that have nothing to do with the quality of the underlying algorithm. The training data becomes stale as the portfolio evolves. The portfolio composition shifts toward product types the […]
