Rocket Close Enhances Mortgage Document Processing with AWS

Rocket Close, a Detroit-based title and appraisal management company, has transformed its mortgage document processing by turning a time-consuming manual task into an efficient automated solution. Processing around 2,000 abstract package files daily, each averaging 75 pages, the company faced a significant operational challenge: manual extraction took an average of 10 hours per package, creating considerable resource allocation burdens and workflow bottlenecks.

Through a partnership with the AWS Generative AI Innovation Center, Rocket Close developed an intelligent document processing solution that has significantly reduced processing time, making the process 15 times faster. The solution utilizes Amazon Textract for OCR processing and Amazon Bedrock for foundational models, achieving a strong 90% overall accuracy in document segmentation, classification, and field extraction.

Amazon Bedrock is a fully managed service that provides a secure way to build and scale generative AI applications. It offers a single API to access leading foundational models from various AI companies. Designed to scale to over 500,000 documents annually, this transformation positions Rocket Close at the forefront of technological innovation in the mortgage industry, supporting faster customer service and sustainable business growth.

Rocket Close handles a high volume of complex documentation as part of its title and appraisal management services, dedicated to helping clients achieve their dream of homeownership and financial freedom. By analyzing a wide range of data points, Rocket Close can quickly and accurately assess the risk associated with loans, enabling more informed lending decisions.

Among the challenges faced by Rocket Close were volume overload, time-intensive workflows, and high financial costs. Manual processes could not keep pace with growing demand, leading to quality concerns and inaccuracies in data extraction. Recognizing the need for a solution that maintained accuracy while dramatically reducing processing time, Rocket Close embarked on developing a new approach.

Abstract document packages are comprehensive collections of legal documents related to property ownership and transactions. These packages typically contain 50 to 100 pages of various document types, often with inconsistent formatting and complex structures, posing unique challenges for automated processing.

As a result of the collaboration between the AWS and Rocket Close teams, a solution was developed that leverages generative AI capabilities to automate the abstract package processing workflow, significantly enhancing efficiency and accuracy.