Customer, Industry, Product, Use Case

Customer: Aditya Birla Finance Ltd. (ABFL)
Industry: Non-Banking Financial Company (NBFC) - Retail & SME Lending (₹35,000 Cr+ AUM)
Product(s): Tartan HyperVerify & Risk Decisioning Platform
Primary Use Case: Unified data ingestion and automated creation of Credit Appraisal Memos (CAMs) and underwriting workflows

ABFL underwrites thousands of retail and small-business loans every month. Historically, CAM generation and verification workflows were heavily manual, fragmented across tools, and did not scale with growth - undermining both operational efficiency and borrower experience.

Executive Summary

Under ABFL’s legacy processes, credit risk teams spent 2–3 hours per Credit Appraisal Memo (CAM) - manually gathering bank statements, extracting financial ratios, performing policy checks, reconciling data across spreadsheets, and communicating via email. This manual burden created inconsistent outputs, slow turnaround times, and operational risk that grew with volume.

With Tartan’s Risk Decisioning Platform and HyperVerify APIs, ABFL implemented a unified verification and decisioning layer that automated document ingestion, opened structured data extraction, and auto-generated audit-ready CAMs within minutes. These platforms also surfaced explainable risk signals such as PD/LGD scores, policy breach flags, and standardized financial ratios directly into the Loan Origination System (LOS).

The outcome was transformational:

• 70% reduction in underwriting cycle time

• CAM generation cut from hours to <40 minutes per case

• 16,000+ financial documents auto-parsed every month

• First-cycle delinquency improvements (~18 bps)

• Audit readiness time compressed from 10 days to 2 days quarterly

• Customer-facing turnaround improved from 4 days to 6 hours

• ~14% lift in NPS (borrower satisfaction)

All without compromising risk controls or compliance readiness.

This was not merely automation - it was a shift from fragmented workflows to reliable, scalable underwriting infrastructure.

The Strategic Challenge: Manual Workflows at Scale Are Fragile

ABFL’s underwriting teams were constrained by the fact that key credit decisions relied on analysts manually piecing together data across disparate sources:

• Bank statements, GST filings, payroll slips, and financial ratios were pulled across multiple tools.

• Each CAM required extensive copy-paste, manual calculations, and analyst coordination.

• Compliance evidence and regulator-ready audit trails were not systematically captured.

• Turnaround times directly scaled with loan volume.

The manual nature of the work meant that increases in application volumes translated into non-linear increases in cycle times, operational cost, and risk of error - conditions that degrade competitive positioning and borrower satisfaction.

Strategic Insight: In credit operations, manual synthesis of data is an operational tax that compounds with growth. Institutions that don’t automate verification and decision genesis trade speed - and ultimately revenue - for avoidable internal effort.

The Solution: Unified Verification + Intelligent Decisioning

Tartan’s solution approached the problem systematically by collapsing multiple steps into a single, API-driven infrastructure layer:

1. Automated Data Ingestion: HyperVerify’s unified APIs ingest bank statements, GST data, payroll slips, and other document types within seconds - a process that previously required human curation.

2. Structured Extraction with OCR & NLP: Tartan’s OCR and NLP engine automatically extracts 1,800+ fields per borrower, normalizing them into structured financial variables without manual work.

3. Risk Decisioning Platform: Instead of analysts manually creating CAMs, Tartan’s platform auto-generates them, embeds explainable risk scores like PD/LGD, and flags policy breaches directly into ABFL’s LOS via a single React component - eliminating swivel-chair operations.

4. Embed in Existing Systems: Critically, this unified decisioning layer slides into existing workflows without requiring a large platform overhaul - accelerating implementation and reducing execution risk.

Insight: Automated underwriting workflows eliminate not only manual toil but also the decision latency that occurs each time a human operator waits on data or context.

Operational Impact – Velocity, Confidence, and Scale

Post-deployment, ABFL’s underwriting workflows were reshaped:

• Analyst hours per CAM dropped from 2–3 hours to <40 minutes, freeing up risk personnel for higher-value exceptions handling.

• Monthly document parsing expanded from ~3,500 to 16,000+, supporting scale without proportional cost.

• Audit readiness time for regulators compressed from ~10 days/quarter to ~2 days/quarter, significantly reducing regulatory work effort.

• Customer-facing underwriting turnaround decreased from ~4 days to ~6 hours, dramatically improving time-to-funding and borrower experience. 

Operational velocity became predictable rather than a function of volume spikes - a core requirement for control-oriented finance functions.

Insight: Predictability is a fundamental aspect of scale. Manual processes can be fast occasionally, but only automated, API-driven systems deliver consistent performance under pressure.

Business Impact – Experience, Risk, and Competitive Edge

The business impact extended far beyond internal efficiency:

• Faster underwriting increased funded volume without additional staffing costs.

• Improved borrower experience drove a measurable lift in NPS, making ABFL stickier at a time when customer loyalty is hard to earn in lending.

• Structured risk outputs (PD/LGD, policy flags) enabled better portfolio performance and reduced early delinquencies (~18 bps).

• Regulatory preparedness improved, with audit logs and structured data captured at each step.

In markets where borrowers equate timeliness with trust, these advantages directly translate into competitive positioning and revenue growth.

Insight: Operational acceleration that preserves risk discipline becomes a strategic differentiator - not merely a tactical gain.

Customer Perspective

“Integrating Tartan’s unified APIs took less than a sprint and eliminated weeks of manual spreadsheet work.”
- Rajesh Shetty, Chief Risk Officer, ABFL

“Our analysts focus on exceptions now - the AI handles the grunt work and produces an audit-ready CAM every time.”
- Divya Narain, VP Credit Ops, ABFL

These voices illustrate a shift in role function - from reactive, manual data processing to strategic risk oversight and portfolio management.

What This Enables Going Forward

With a unified risk-decisioning layer in place, ABFL is positioned to:

• Scale underwriting throughput without linear increases in analyst hours.

• Deploy early-warning triggers for collections using the same automated data layer.

• Extend automation into other credit products (e.g., commercial loans, dealer financing).

• Introduce advanced analytics that leverage structured data for pricing, segmentation, and loss forecasting.

Executive Insight: When verification and decisioning infrastructure become reliable building blocks, organisations unlock growth optionality - the ability to pursue adjacent products, enter new markets, and run portfolio experiments with minimal risk.

About Tartan

Tartan provides unified APIs and intelligent decisioning platforms that turn manual, fragmented processes into automated, scalable infrastructure - empowering enterprises in lending, insurance, banking, and fintech to accelerate operations while preserving compliance and control.