Insights

Customer Success Stories

Casestudy

AI enablement of Loan2Wheels Platform

Initial intervention areas for AI enablement

1. On-demand Scalable AI Bots for Quality check bucket and entire Backend Operations

  1. Will perform Partial or Full process activities at each of the operations bucket
  2. Will assign activities automatically to itself and process them as assigned to an agent
  3. Cleared tasks by AI bot can be taken by Human Agents for processing remaining activities
  4. During peak demand AI bots can Auto-scale in numbers to pick up extra load

2. AI powered Audits

  1. Will review Active Loans to verify compliance and create NC(Non Compliance) reports
  2. Will review defaults and closed loans to identify patterns of possible issues
  3. The coverage of audit could be 100% rather than sample loans only

3. Gen AI powered digital loan assistant for Marketing and Lead Gen

  1. Will chat with applicants to understand their needs
  2. Recommend best possible 2-wheeler option and corresponding loan
  3. Based on the Documents / Info shared create in-principle approval
  4. Create Lead data for CRM

AI Future Roadmap for Loan2Wheels

1. Loan Portfolio Monitoring and Risk Management :

Continuously monitor loan portfolios and apply Predictive analytics through regression and decision tree models that use patterns observed in historical data to identify borrowers with a high tendency to default across segments

This can be achieved by creating Business Intelligence based on a data lake from Loan Management System (LMS)

Advantage – identify potential delinquencies, and suggest risk mitigation strategies, ensuring prudent risk management practices.

2. Collection analytics

Collection analytics to streamline its collection process. Using data points such as gross amount owed and contact attempts, an ML model can be trained to identify a debtor’s propensity to pay.

Advantage – Robust collection processes with lower operating cost and higher efficiencies.

3. Field Visit Optimization with Intelligent Field Visit Scheduling:

Analyze customer data and past loan applications to prioritize field visits and schedule loan officers efficiently, optimizing resource utilization

Advantage – Reduced cost and time of field visits of agent to focus on high quality applicants first.

4. Eliminate Biases from underwriting–

Human bias would frequently seep into the underwriting process, causing application rejections or higher/lower interest rates to be applied to some loans, which would result in consumer displeasure or loss for the lender. ML algorithms will look for patterns and give insights and conclusions based on the same to eliminate this bias. This can lower error rates and produce loans that are more profitable for lenders

Advantage - Higher profitability, operational efficiency and adherence to RBI framework for micro finance institutions

Detailed processes for AI enablement

AI enabled Quality Check Bucket

Process Bucket AI Bot activities Ops Team activities Business Outcome / Benefits
Quality Check

Before calling IDFY for ID verification, AI Bots can

  1. Match Photos Across Documents with Photo taken or uploaded by Sales Agent
  2. Match Name, and other common Info across Documents to eliminate Inconsistencies

Computer Vision assisted Profile Photo Quality check

Automate the verification of KYC documents for Facial recognition and document analysis against Govt Databases, ensuring compliance with regulatory requirements.

 Will only focus on Documents cleared by AI bot having no data or photo inconsistencies
  1. Lower Turn Around Time(TAT) for Field Agents so that they can sell more during peak time like Festivals etc.
  2. Lower errors and cost saving by not calling IDFY API for inconsistent documents
  3. Better Customer experience and buying decision with faster response

AI Bots for entire Backend Operations

Process Bucket AI Bot activities Ops Team activities Business Outcome / Benefits
Eligibility Check

Highlight Applications as RED/AMBER/GREEN for an applicant.

AI bot will determine exposure limit using ML model based on various factors such as income, spending habits and credit history, demographic profile, Item possessions etc.

 Will scrutinise more for RED and AMBER marked applications to create appropriate exposure limit and loan value
  1. Avoid possible errors while making credit limit decisions specially during peak period
  2. Reduced NPA and higher profits per customer
Sanction Maker

Field Visit Report Analysis: Analyze field visit reports, extract key information, and identify potential red flags, facilitating informed final loan decisions.

Analyse Applicant’s data for possible errors or issues.

 Will scrutinise more for RED flagged applications before sanctioning the loan
  1. Reduced cost and higher efficiency through lower TAT for loan processing/decision making
  2. Reduced NPA

AI Bots for entire Backend Operations

Process Bucket AI Bot activities Ops Team activities Business Outcome / Benefits
Disbursement Maker & Disbursement Check

Verify and match the data in Images & text files with the filled up text data e.g. Quotation, IDs, Address , FI report etc.

Highlight all the possible issues as per set business rules.

 Will scrutinise more for RED flagged applications before disbursement
  1. Reduced errors of manual intensive process of verification
  2. Reduced NPA
Post Disbursement & Registration No Check

Verify and match

  1. Original Invoice and Insurance with Data on System
  2. Image of RC with RC data from System

Highlight the possible issues

 Will focus more on RED flagged applications

Gen AI powered digital loan agent

Process Bucket AI Bot activities Ops Team activities Business Outcome / Benefits
Marketing
  • Will enable conversation with customers or referrer, to understand their needs with help of conversational AI
  • Recommend best possible 2-wheeler option with corresponding loan
  • Based on the Documents/Info and preliminary risk assessments, create in-principle approval before even visiting the Showroom.
  • Create Marketing Qualified Leads in CRM
  • - Identify prospects who did not complete the process and record them as targeted audience for Marketing campaigns
 Will only focus on completing the remaining Loan Application processes
  1. Faster conversion and sale
  2. Creation of large potential leads as top of the marketing funnel.
  3. Better Customer experience and buying decisions

AI powered automated loan audits and compliance monitoring

Process Bucket AI Bot activities Ops Team activities Business Outcome / Benefits
Audit
  1. Review the Loans for compliance checks at predefined Monthly or Quarterly frequency
  2. Create NC reports and possible Root Cause Analysis ( RCA)
  3. Identify NC Patterns over the period of time to create possible mitigation strategies
 Will check the AI Generated NC reports as well to enforce mitigation plan.
  1. Full coverage of compliance across all the loans
  2. Identification of areas of concerns having maximum issues
  3. Improved reporting

Implementation Roadmap

Short term :

  1. AI powered Quality Check bucket
  2. On-demand Scalable AI Bots for Backend Operations (Initially Decision Support System which can evolve to Full-Automation level)
    1. Credit Score and Exposure recommendations in Eligibility Check bucket
    2. Analysis of Field Inspection reports and Auto check of Applicant’s Info for Sanction Maker bucket
    3. Data Verification in Disbursement and Post Disbursement check bucket
  3. AI powered automated loan audits and compliance monitoring

Medium Term :

  1. Gen AI powered digital loan assistant for lead-gen through automated in-principle approvals
  2. AI led verification process for Post Disbursement Bucket
  3. Field Visit Report Analysis for Sanction Maker bucket

Long Term :

Refer AI Future Roadmap for Loan2Wheels

Business benefits and outcomes

- Automated Quality check bucket and On-demand Scalable AI Bots for Backend Operations

For Sales team : Near real-time credit decisions for field agents to sell more loans during peak periods

For Customers : Faster TAT for customer for faster buying decision against competition

For Ops team : Auto-scaling of back-end operations for varying demand of high and low volumes

For Business

  1. Cost savings by reducing rework due to data errors and inconsistencies
  2. Cost savings by avoiding third party pay-per-use solutions with wrong/inconsistent data
  3. Reduced NPA and higher profits per customer

AI powered automated loan audits and compliance monitoring

For Compliance team

Review of all the loans disbursed to highlight issues rather than only sample audits

For Business

Lower NPA by identifying issues early

Business benefits and outcomes

- Gen AI powered digital loan assistant for Marketing and lead-gen

For Sales team
  1. Faster conversion with customers having in-principle approvals before turning up at store
  2. Lead Gen tool for creating top of the marketing funnel
For Customers
  1. Better Experience with enough time for making a buying decision