How do Retail Lending Companies Automate Loan Origination Processes Using GenAI?

The retail lending industry is growing worldwide. In the United States, private sector loans increased from $2.5 billion in February 2022 to $2.77 billion in July 2024. Indian NBFCs (Non-Banking Financial Companies) are also not far behind. The lending market share in consumer durables stood at 61% in 2023.

That’s a clear win for retail lending companies, right?

Well, not really. This growth has also resulted in a high cost of loan origination. Investing and maintaining technology, managing multiple integrations, and paying high remuneration costs have led to a financial crisis for the global retail lending industry.

However, the good thing is that this industry is taking active steps toward digital adoption.

When discussing digital adoption, how can we skip AI? AI can potentially transform the loan origination process through automatic data capture and extraction, personalized support, income verification, and more.

In this article, we will perform a before-vs.-after scenario analysis of retail lending companies to assess how AI can make a significant difference.

Let’s get started!

‍

The Current Load Origination Process

Currently, the loan origination process at a broad level looks something like this 👇

‍

‍

#1 Pre-qualification stage

The first stage of any loan origination process is to take the borrower through a pre-qualification process. In this stage, the borrower must share all relevant documentation, such as identity proofs, income certificates, employment details, tax expenses, loan amount, purpose of loan, etc. The lending company performs due diligence on these documents to ensure the buyer qualifies to move on to the next stages.

#2 Application and documentation stage

Once the lender notifies the borrower about pre-qualifying for the loan, the next step is to apply for the loan with proper documentation. This stage involves lots of paperwork. Some lenders still prefer to do this manually, while many have shifted to digital documentation. In this stage, the borrower needs to submit all information and documents to reach the application processing stage.

#3 Application processing stage

After all documentation and application forms are submitted, the processing stage begins. In this stage, the lending company reviews the credibility of all documentation and analyzes the borrower’s credit profile to ensure they can repay the loan on time.

#4 Underwriting stage

The underwriting stage is the most critical one in any loan origination process. It is when a commercial underwriter analyzes multiple factors such as financial statements, credit history, debt-to-income ratio, industry trends, assets, and collaterals of the borrower. By the end of this stage, the lending company decides whether to approve a loan and the associated credit risks.

#5 Credit decision stage

In the decision stage, the lender communicates its final decision to the borrower based on their background and credit profile analysis. The decision can be loan approval, rejection, or even a lower loan amount.

#6 Quality check stage

Even though your loan application is approved, it goes through a final quality check stage. This stage focuses on local compliances and regulations.

#7 Loan approval and disbursement stage

Once the borrower and lender agree to the terms and conditions of the loan, amount, and repayment, the loan amount is disbursed to the borrower. The mode of payment can be cheque, bank transfer, or any other pre-decided mode. The entire process can take up to 60 days.

#8 Post-approval servicing stage

Then the post-approval servicing stage starts, in which the lending company supports the borrower regarding their loan account, payments, and customer support.

‍

The Potholes of the Current Loan Origination Process

The biggest pothole in the current loan origination process is that it is manual. In most cases, the borrower must visit a branch to apply for a loan and interact with a human agent to understand the process.

Credit analysts and underwriters gather all information related to the borrower’s credit profile. Finally, the underwriter manually analyzes vast amounts of data to decide whether to approve or reject the loan.

Even after a loan is approved, the borrower is assigned a loan assistant who manually completes all loan documentation before disbursing the loan.

No wonder all these manual processes and systems lead to:

• High overhead cost as you need to hire multiple employees to drive different stages of the loan origination process
• Long, exhaustive, and manual assessments lead to errors in loan decision-making
• With so much paperwork involved and due to lack of security, losing sensitive customer information is very common

“A common issue raised with many client organizations is the lack of care that is apparent among lenders. In most of our client organizations, the document preparation function for commercial loans usually resides in the loan operations function. Lenders regularly complain that the document prep process takes entirely too long. It may be several days or longer before a document arrives back ready for a borrower’s signature. The primary driver of this delay often stems from errors created early in the process by lenders, loan assistants, or credit analysts that document prep staff catch before preparing the final loan documents.” - John Mateker, Vice President at Ceto and Associates.

‍

How can Conversational AI Change it on all Touchpoints?

A study by Accenture shows that 59% of banking employees already use AI every day. Lending companies should make AI part of their workflows. A PWC study showed that 57% of respondents in the financial sector strongly agree that AI improves operational efficiency.

‍

Source

‍

AI helps lenders automate all repetitive processes in the current loan origination process, making loan processing faster and more accurate.

Here is how the loan origination process will look like with the help of GenAI implementation:

#1 Application submission stage

Currently, loan application submissions are mostly done manually. Even though the online application option is available, it involves downloading the online form and submitting it physically at the branch.

However, the process is completely different and faster with AI in the picture. GenAI helps borrowers fill out the loan application form in real time. It answers any questions the applicant may have during the process and offers multilingual support to make the application submission stage more inclusive.

It helps retail lending companies increase the efficiency of the application stage, automate the stage without any human agents in the loop (unless special scenarios occur), and reduce manual errors in the application process.

For borrowers, AI's involvement leads to a personalized experience, where they receive assistance throughout the application submission stage. AI also helps the borrowers process loan applications faster (think minutes instead of days).

Metrics impacted:

• Processing time: It reduces to minutes instead of days or weeks
• Customer satisfaction level improves

#2 Data entry stage

The data entry stage involves staff manually entering application data into the system. This process is time-consuming, prone to human errors, and requires significant human resources.

AI-powered systems can now automatically capture data from digital applications, extract relevant information, and validate it in real-time. This eliminates the need for staff to manually enter data.

For retail lending companies, AI-driven data entry reduces operational costs as fewer staff members are needed. The accuracy of the data improves significantly, reducing errors that could lead to issues later in the loan process. This automation allows companies to process more applications with incredible speed and precision.

For borrowers, the AI-powered data entry stage means reduced documentation requirements. They no longer need to provide as much paperwork as the system can extract and verify information from digital sources, resulting in a smoother, less burdensome application process.

Metrics impacted:

• Operational Costs: Decrease due to reduced need for manual data entry staff
• Accuracy Rate: Increases as AI reduces human errors in data entry

‍

Source

‍

#3 Credit analysis stage

The credit check stage involves staff manually requesting and reviewing credit reports for loan applicants. This process is time-consuming and can lead to delays in loan processing.

The credit check process becomes instant and more comprehensive with an AI-powered system. AI-powered systems perform instant credit checks by seamlessly integrating with credit bureaus and other financial data sources. The system can analyze not only credit reports but also income statements, tax returns, and other financial data to provide a more holistic view of the applicant's creditworthiness.

For retail lending companies, AI-driven credit checks lead to faster processing times and improved decision-making. The system can quickly analyze vast amounts of data, identifying patterns and risk factors that human analysts might miss. This results in more accurate risk assessments and better loan decisions. The AI can even provide real-time loan decisions for retail and small-ticket loans, significantly reducing processing times and operational costs.

For borrowers, the AI-powered credit check stage means instant decisions and reduced wait times. They benefit from a more thorough analysis of their financial situation, which could lead to fairer lending decisions. The process is less intrusive as the AI can gather much of the required information without additional documentation from the borrower.

Metrics impacted:

• Processing Time: Drastically reduced from days to minutes or seconds
• Approval Rate: Potentially improved due to more comprehensive analysis
• Risk Assessment Accuracy: Increased due to AI's ability to analyze multiple data sources

#4 Income verification stage

The income verification stage involves staff manually verifying income through various forms of documentation provided by the applicant. This process can be time-consuming, labor-intensive, and prone to human error.

With AI in your workflow, income verification becomes more automated and sophisticated. AI-powered systems can now analyze digital documentation and various data sources to verify income quickly and accurately. This includes analyzing pay stubs, bank statements, tax returns, and non-traditional income sources.

For retail lending companies, AI-driven income verification improves accuracy and reduces manual errors. The system can cross-reference multiple data points to ensure the validity of the income information, potentially uncovering discrepancies that human reviewers might miss. This results in better risk assessment and more informed lending decisions. The automation also allows for faster application processing, increasing overall efficiency.

For borrowers, the AI-powered income verification stage means a faster verification process and reduced documentation requirements. They may not need to provide as much paperwork as before, as the AI system can often verify income from digital sources they authorize access to. This results in a more convenient and less burdensome application process.

Metrics impacted:

• Accuracy Rate: Increases as AI reduces human errors in income verification and can detect anomalies more effectively
• Processing Time: Decreases significantly as AI can verify income much faster than manual processes

#5 Identification verification stage

The identification verification stage involves staff manually verifying the applicant's identity through various forms of documentation. This process can be time-consuming, susceptible to human error, and potentially vulnerable to fraud.

AI integration helps digitize the identification verification process, making it automated and more secure. AI-powered systems employ biometric authentication and digital ID verification techniques to confirm an applicant's identity quickly and accurately.

For retail lending companies, AI-driven identification verification enhances security and reduces fraud risk. The system can use advanced technologies like facial recognition, fingerprint scanning, or even behavioral biometrics to ensure the person applying for the loan is who they claim to be. This significantly reduces the risk of identity theft and fraudulent applications. The automation also speeds up the verification process, improving overall efficiency.

For borrowers, the AI-powered identification verification stage offers a convenient digital experience with reduced documentation requirements. Instead of providing multiple forms of physical ID and waiting for manual verification, they can often complete the process quickly through their smartphone or computer. This might involve taking a selfie for facial recognition, scanning their fingerprint, or using other biometric identifiers.

Metrics impacted:

• Security Risk: Decreases due to advanced biometric authentication methods
• Customer Satisfaction: Improves due to the convenience and speed of digital verification
• Processing Time: Reduces significantly as AI can verify identity much faster than manual processes

Source

‍

#6 Underwriting stage

Underwriting involves staff manually reviewing the application, credit report, and other factors to assess the loan's risk and determine approval. This process can be time-consuming, subjective, and potentially biased.

AI-powered systems evaluate creditworthiness using many data points and sophisticated algorithms, providing a more comprehensive and objective assessment.

For retail lending companies, AI-driven underwriting leads to improved decision-making and reduced bias. The system can analyze traditional factors like credit scores, income, and alternative data sources to create a more holistic view of the applicant's financial health. It can also automate the creation of key underwriting documents, including executive summaries and business descriptions.

This automation speeds up the process and ensures consistency in evaluations across applications. The AI can also continuously learn from loan performance data, refining its models to improve accuracy.

For borrowers, the AI-powered underwriting stage means faster decisions and reduced wait times. The comprehensive nature of AI analysis may also lead to fairer lending decisions, as it can consider a broader range of factors beyond traditional credit scores. This could open up lending opportunities to borrowers who conventional underwriting processes might overlook.

Metrics impacted:

• Approval Rate: May increase due to more comprehensive analysis of borrower creditworthiness
• Processing Time: Significantly reduced as AI can analyze data and make recommendations much faster than human underwriters
• Consistency: Improves as AI applies the same criteria uniformly across all applications

#7 Decision stage

Currently, the decision stage involves staff deciding on loan approval or denial, often based on subjective criteria and the underwriter's analysis. This process can be time-consuming and may lead to inconsistent outcomes.

The decision process becomes more data-driven in the presence of AI. AI-powered systems can now make instant decisions on loan applications, especially for retail and small-ticket loans, based on a comprehensive analysis of all available data.

For retail lending companies, AI-driven decision-making leads to increased efficiency and improved consistency. The system can process applications and make decisions 24/7, significantly reducing turnaround times. It applies consistent criteria across all applications, reducing the potential for bias or inconsistency with human decision-makers. This automation allows companies to handle more applications without proportionally increasing staff.

For borrowers, the AI-powered decision stage means instant decisions, reducing uncertainty and wait times. They can often receive an immediate response to their loan application, allowing them to plan accordingly. The consistency of AI decision-making also contributes to a fairer lending process, where similar applications are more likely to receive similar outcomes.

Metrics impacted:

• Approval Rate: May be optimized as AI can more accurately assess risk and creditworthiness
• Customer Satisfaction: Improves due to faster decisions and reduced uncertainty
• Processing Time: Drastically reduced from days or weeks to minutes or even seconds

#8 Documentation stage

The documentation stage involves staff manually preparing and sending loan documents for signing. This process can be time-consuming, prone to errors, and often requires physical paperwork.

AI-powered systems can now generate and send documents for e-signature, streamlining the entire process.

For retail lending companies, AI-driven documentation reduces operational costs and improves efficiency. The system can automatically generate accurate loan documents based on the approved terms, reducing the risk of errors that can occur with manual document preparation. This automation allows for faster processing of a higher volume of loans without a proportional increase in staff. The digital nature of the process also reduces costs associated with printing, mailing, and storing physical documents.

For borrowers, the AI-powered documentation stage offers a convenient digital experience with faster processing. Instead of waiting for physical documents to arrive by mail, reviewing them, and returning signed copies, borrowers can receive, review, and sign documents electronically, often from their own devices. This significantly reduces the time from loan approval to fund disbursement.

Metrics impacted:

• Operational Costs: Decrease due to reduced need for manual document preparation and handling
• Processing Time: Significantly reduced as document generation and signing can happen almost instantly
• Customer Satisfaction: Improves due to the convenience and speed of digital documentation

#9 Closing and loan servicing stage

The closing stage involves customers signing documents and disbursing funds, often requiring in-person meetings or manual processing. Loan servicing typically involves human agents handling customer inquiries, managing accounts, and processing payments.

AI-powered systems can digitally manage the closing process and provide intelligent, proactive loan servicing.

For retail lending companies, AI-driven closing and loan servicing improves customer experience and reduces manual errors. The system can automatically disburse funds upon e-signature completion, eliminating delays.

In the servicing phase, AI can handle routine inquiries, process payments, and even predict and prevent potential issues before they arise. This automation allows companies to manage a larger loan portfolio without a proportional increase in staff.

For borrowers, the AI-powered closing and loan servicing stage offers faster funding and a more personalized, responsive service experience. They can complete the closing process quickly and conveniently, often from their devices. During the loan term, they benefit from:

• Automated payment reminders tailored to their preferences
• Instant responses to billing inquiries
• AI-assisted account management
• Personalized recommendations for loan repayment based on their financial history and behavior

Metrics impacted:

• Customer Satisfaction: Improves due to faster funding and personalized, responsive service
• Processing Time: Reduced as fund disbursement happens automatically upon e-signature
• Operational Efficiency: Increases as AI handles routine servicing tasks
• Default Rates: Potentially decrease as AI provides proactive, personalized payment reminders and repayment recommendations

‍

Role of Conversational AI in the Loan Origination Process

Conversational AI streamlines the loan origination process. Let’s understand with a case study. A PWC survey shows that chat automation is the highest-implemented AI use case across organizations, with an 82.6% response rate.

Source

The organization is a leading digital lending company with $1B+ AUM. By implementing an AI-powered assistant by Alltius, the lender drastically reduced customer support wait times from 4 hours to just 5 minutes while handling over 3,000 emails daily.

Here’s what the lending company was struggling with:

• Customers faced 4-5 hour waits for responses, primarily due to pre-opening hours ticket queuing and assignment delays
• Response quality varied based on individual agent's tenure and knowledge
• New agents required 3-6 months to become as effective as experienced ones
• The lender received over 10,000 tickets daily across various channels
• Tailoring responses to customers with different account profiles, default statuses, and credit scores was challenging

Alltius addressed these challenges by taking the following actions:

• Implemented an AI assistant trained in past conversations and company knowledge
• Integrated the AI with existing customer support platforms
• Used AI to generate personalized draft responses with placeholders for sensitive information
• Allowed human agents to review and edit AI-generated responses before sending
• Continuously improve the AI model based on new interactions and feedback

These initiatives resulted in the following:

• The response time was reduced from five hours to two minutes
• There was a 2X increase in contact center productivity
• The training time and agent ramp-up time decreased from six months to one month

‍

Conclusion

Integrating Conversational AI in the loan origination process marks a significant leap forward for retail lending companies. GenAI reduces processing times, improves accuracy, and enhances customer satisfaction by automating various stages- from application submission to loan servicing.

If you are looking for a way to automate your lending origination process, start with a conversational AI platform like Alltius. It is a good starting point for automating loan origination stages like application submission and support.

Want to explore how it can help in your loan process automation journey?

Book a demo!

‍