Consumer Finance Company Achieves Growth Using Artificial Intelligence (AI) With Microsoft Azure Cognitive Services

Consumer Finance Company (CF) decided to undergo a digital transformation to drive growth. The priorities were to increase conversion, stop fraud and decrease costs. 

Problem

To improve conversion, CF created an end-to-end digital lending platform which would offer a great customer experience with significantly lower Turn Around Time (TAT). Artificial Intelligence (AI) used in the on-boarding process would replace manual steps (thus lowering operational costs) and would keep fraud in check.

Context

CF’s legacy loan application process was manual - both on the customer side and the operations side. The process steps and problems are explained as below.

Step 1: Loan-seeker fills application and submits documents

Manual filing of forms and document submission was prone to errors. Tech savvy customers found the process cumbersome and archaic.

Step 2: CF staff assesses the application manually

Checking documents manually and field verification took time and was costly and inefficient, and allowed some fraud to pass.

Step 3: Paper contract is signed and loan amount is disbursed

TAT was around 4-5 days causing poor customer experience and high dropouts.

Solution

Know Your Customer (KYC) 

Integrate Microsoft Azure Cognitive Services and its KYC technology into CF’s new workflow as below:

Azure Cognitive Services


Step 1: KYC Starts

Consumer starts a new application and reaches the KYC point.

Step 2: Identity Document (ID) Card

Customer uploads a photo of his/her ID card. Microsoft Azure Cognitive Services Vision technology checks the image for errors (e.g.: blurred image) and fraud (e.g: photo pasted) and asks the customer to re-upload if required.

Step 3: Selfie

Customer takes a selfie (no need to shake head or blink or smile) to prove liveness. Microsoft Azure Cognitive Services - Vision technology checks the selfie for liveness, checks errors (e.g.: bad lighting) and frauds (e.g: sleeping parent) and asks the customer to retake if needed.

Step 4: Real-Time Check

Microsoft Azure Cognitive Services - Vision technology
(1) matches face in selfie to ID card,
(2) performs a 1:N check with face database for red flags, 
(3) and pre-fills the application form with OCR (Azure's Computer Vision API includes Optical Character Recognition (OCR)) data from ID card.

Vision Bot Architecture

Step 5: KYC Ends

The entire KYC process is completed within minutes, and the next step in the loan application process starts.

Products & Services  

Azure Cognitive Services
A comprehensive family of AI services and cognitive APIs to help you build intelligent apps
https://azure.microsoft.com/en-in/services/cognitive-services/ 

Azure's Computer Vision 
An AI service that analyzes content in images and video
https://azure.microsoft.com/en-us/services/cognitive-services/computer-vision/ 

Face
An AI service that analyzes faces in images
https://azure.microsoft.com/en-us/services/cognitive-services/face/ 

Form Recognizer
The AI-powered document extraction service that understands your documents
https://azure.microsoft.com/en-us/services/cognitive-services/form-recognizer/

Data Storage And Security

The solution offers clients the option to delete images immediately after processing. It temporarily stores images in a cloud-based No-SQL Database - Cosmos DB. Data encryption at rest and transit is enabled across all resource components of architecture, and the keys are stored securely in the Key Vault.  Therefore, the solution complies with the security and privacy requirements of the Banking, Financial Services and Insurance (BFSI) sector. All Azure resources used are ISO compliant as well.

Products & Services

Azure Cosmos DB
Fast NoSQL database with open APIs for any scale
https://azure.microsoft.com/en-us/services/cosmos-db/ 

Key Vault
Safeguard cryptographic keys and other secrets used by cloud apps and services
https://azure.microsoft.com/en-us/services/key-vault/

Impact

The new onboarding system provides a quick customer experience, and it prevents fraud at the acquisition step, saving hundreds of thousands of dollars

The next problem that Microsoft Azure Cognitive Services is solving for CF is to create a multi pronged approach to assess risk and identify fraud across all their products using AI.

References

KYC in banking
https://docs.microsoft.com/en-us/common-data-model/schema/core/applicationcommon/foundationcommon/crmcommon/accelerators/financialservices/banking/kyc

Transforming the ‘Know Your Customer’ process using Optical Character Recognition
https://www.microsoft.com/en-in/msidc/blogs/Articles/Transforming-the--Know-Your-Customer--process-using-Optical-Character-Recognition/513c3191-2245-4587-aab1-fb45032d51d7 

Azure Cognitive Services pricing
https://azure.microsoft.com/en-us/pricing/details/cognitive-services/


Comments

Let's connect. A great way to get my attention is to comment on one of my posts.

Get In Touch

Send