Non banking financial companies finance growth across retail and MSME segments, yet their credit decisions must balance speed with prudence. A clear, repeatable underwriting approach helps lenders screen risk, price loans fairly, and protect portfolio quality through cycles. This guide maps the Top 10 Credit Underwriting Frameworks for NBFCs that practitioners can adapt for unsecured and secured products. Each framework explains the goal, core levers, and practical checks, so teams can align policies, models, and frontline judgment. The emphasis is on scalable methods that blend data, human oversight, and governance to deliver faster approvals without compromising control.
#1 Risk based lending with PD, LGD, and EAD
Anchor risk based lending on probability of default, loss given default, and exposure at default. Define risk grades from A to E tied to PD bands, then set pricing, limits, and approval levels by grade and collateral. Use bureau data, income surrogates, bank statement trends, and sector outlook to estimate PD, and collateral haircuts to estimate LGD. Document override rules and require second line review for material exceptions. Update grade migration matrices quarterly to capture cycle turns and concentration shifts. With this framework, frontline teams can decide quickly while the firm keeps capital, provisioning, and pricing consistent with expected losses.
#2 Cash flow centric underwriting for capacity and survivability
Center decisions on verified cash flows rather than static income proofs. Parse bank statements to build monthly inflow and outflow patterns, seasonality, and volatility. Compute survivability after obligations using FOIR or debt to income, but stress cash flows with realistic shocks, such as revenue dips and expense spikes. Cross validate invoices, GST data, and POS feeds for MSME borrowers. Use cash flow models to propose tenor, repayment frequency, and installment structures that match working capital cycles. Require enhanced scrutiny where cash relies on a single counterparty. This framework improves inclusion while protecting repayment capacity under adverse conditions.
#3 Dual scorecard approach using bureau and application models
Combine bureau scorecards with application scorecards for a robust two dimensional risk view. Segment by vintage, product, geography, and employment type, then calibrate cutoffs and pricing by segment. Use reject inference to correct biases and recover signal lost from historical declines. Monitor characteristic stability indices to detect drift in variables like recent delinquencies or enquiry intensity. Enable strategy trees that route low risk cases to straight through processing, medium risk to assisted approval, and high risk to decline or additional verification. Retrain quarterly or when Gini drops materially. This framework scales volumes without diluting risk discrimination or fairness across cohorts.
#4 Alternative data and bank statement analytics
Augment traditional data with alternative sources that sharpen early risk detection. Use bank statement analyzers to extract bounce rates, minimum balance dips, end of month sweeps, and salary credit stability. Blend telecom, utility, and e commerce signals where regulations permit and customers consent. Apply feature governance to log data origin, transformation, and privacy justifications. Prioritize variables that are stable, interpretable, and portable across segments. Backtest uplift versus a bureau only baseline and retain only features that add incremental Gini. This framework enhances thin file approvals while preserving transparency, auditability, and respect for customer privacy and consent.
#5 Collateral valuation, LTV discipline, and enforceability
For secured lending, build a disciplined collateral framework that links valuation, legal enforceability, and liquidation timelines to limits. Set product level loan to value caps that vary by asset class, geography, asset age, and borrower segment. Use independent valuers, revaluation triggers, and conservative haircuts for volatile assets. Validate title, encumbrances, and insurance coverage before disbursal, with clear loss payee clauses. Define step up covenants, inspection frequencies, and release conditions tied to repayment behavior. Stress LGD using forced sale discounts and time to recover. This framework protects downside, supports risk based pricing, and aligns security realization with expected recoveries.
#6 Income verification and affordability metrics
Standardize income verification with graded evidence rules that vary by profile. For salaried borrowers, require salary credits, employer checks, and tax statements. For self employed borrowers, triangulate GST filings, bank flows, audited accounts, and cash flow surrogates. Compute affordability using FOIR, debt to income, and fixed obligation coverage, then cap obligations prudently by segment. Apply negative filters such as recent cheque bounces or payroll variability. Document exceptions with quantified mitigants, such as co borrower support or higher equity contribution. This framework ensures repayment capacity is demonstrated, verifiable, and resilient to plausible shocks before approving any exposure.
#7 Behavioral scorecards and dynamic limit management
For revolving and repeat credit, deploy behavioral scorecards built on your own performance data. Use payment timeliness, utilization patterns, promise to pay adherence, and complaint signals to predict roll rates. Link score bands to dynamic line assignment, installment amounts, and cross sell eligibility. Set risk based reprice triggers and soft collections nudges for deteriorating segments. Run champion challenger experiments to validate uplift before full rollout. Govern override use by recording decision rationale and outcome tracking. This framework rewards good customers with higher limits and better pricing while curbing exposure to accounts showing stress before delinquency materializes.
#8 Early warning systems and portfolio surveillance
Build an early warning framework that monitors accounts and segments for emerging stress. Track triggers such as bureau delinquency spikes, bounce rate changes, geographic shocks, and sector downgrades. Define traffic light thresholds that map to actions like enhanced contact, review of collateral, or temporary line freezes. Use roll rate matrices and vintage curves to forecast losses and recalibrate collections strategies. Publish monthly dashboards that compare plan versus actual for approval rates, risk grade mix, and expected loss. This framework tightens the feedback loop, allowing policy and model updates before losses escalate, while giving leadership clear visibility on risk appetite adherence.
#9 Governance, policy hygiene, and model risk management
Underwrite within a strong governance framework that defines policies, authorities, and model risk controls. Maintain a single source of truth for underwriting standards, with change logs and training for frontline teams. Establish three lines of defense, periodic audits, and independent model validation covering data, design, and performance. Document intended use, limitations, and monitoring for every scorecard and rule. Use challenger models and backtesting to detect degradation early. Report exceptions and overrides with quantitative impact summaries. This framework reduces operational risk, improves regulatory readiness, and ensures consistency across branches, partners, and digital channels as portfolios scale rapidly.
#10 Scenario based stress testing embedded in decisions
Embed stress testing into underwriting so every approval reflects base and adverse realities. Define macro and sector scenarios that shift PD, LGD, cash flow, and collateral values meaningfully. Run sensitivity for key variables like interest rates, repayment buffers, customer churn, and wage stability. Translate results into approval gates, LTV adjustments, covenants, pricing add ons, or reduced tenors. Refresh scenarios semiannually and after notable shocks or policy changes. Summarize borrower level outcomes in a standardized memo that explains resilience under stress and highlights key mitigants. This framework builds a practical bridge between portfolio risk appetite and frontline decisions, aligning growth with durability across conditions.