North Avenue Capital Automates Credit Eligibility Process With GCP

NAC’s requirement to automate the workflow to assess the credit-worthiness of its customers was successfully achieved with the integration of Google products such as BigQuery, Google Sheets and Forms resulting in process accuracy, customer satisfaction, and a multiplied profit.

North Avenue empowering businesses financially, growing economies and creating employment opportunities in rural America
They dive deep into any business, develop an understanding of the key objectives, and assist in preparing a USDA financing package

North Avenue’s Plan to Automate Data Assessment and the Querying Pipeline

Given the nature of the financial lending industry, a smooth process to evaluate the credit-worthiness should be in place to avoid any complications. North Avenue Capital (NAC) has partnered with various small scale businesses to equip them with debt-financing and thus helping them to execute their business plans. As the organization expanded with more businesses under their portfolio, they faced issues with USDA loan processing and evaluating the credit eligibility of it’s customers. Since the entire process was carried out manually, it added an unnecessary delay in the lead time to process and qualify the loan requirements, thereby, involving significant customer attrition and dissatisfaction.

PPNAC, being a specialized commercial lender of USDA loans, identified this as a bottleneck and made it their near term goal to streamline the system. For this, they partnered with Searce with an objective to automate the entire data assessment and querying pipeline. This was proposed to be achieved by matching the customer address data received to the USDA eligible zone information available. PPP

Searce’s guidance for NAC through Google Cloud Platform (GCP) BigQuery

The team at Searce collaborated with the NAC’s technical team to execute and meet their requirements. The solution proposed by Searce was executed in the following manner:

A Google Form was embedded into NAC’s website that collected customer data including the address
The form was then linked to a Google Sheet wherein each address was converted to the specific geo-coordinates
A Google Sheet configuration continuously transferred the data to Google Cloud Platform (GCP) BigQuery. This was done by converting the shapefile into a BigQuery supported format (GeoJSON fragments) using a Python script
NAC-provided shapefile containing USDA eligible zones was also used with BigQuery
A query to compare the customer geo-coordinates with NAC’s shapefile was run to display the output declaring the customer’s loan eligibility

The case delivery required constant communication between NAC and the Searce team which concluded with a successful implementation of GCP, Google Sheets, and Forms resulting in a solution that could be constructed further by the client’s internal team or with the help of Searce.

The Outcome of Implementing BigQuery in NAC’s Operations

The solution proposed by Searce met the NAC’s immediate requirement. The tailor-made solution had an above the mark result that was observable during the performance of the automated model. A few notable results were:

Elimination or reduction in NAC’s staffing requirements to conduct the manual process
Remarkable reduction in the lead time enabling an almost instant response to the eligibility check
Near to perfect accuracy of the proposed system along with long term benefits to NAC in terms of performance, reliability, and scalability
Accurate assessment of the customers’ risk profile led to an appreciable decrease in NAC’s loan failure rate