Automate Document Processing with Quick Automate and the IDP Accelerator
Classifying, extracting, and validating high volumes of documents is a challenge across banking, insurance, healthcare, and the public sector. See how a mid-size mortgage lender automates its entire document intake pipeline, from email to validated data, using the AWS GAIIC IDP Accelerator and Amazon Quick Automate.
Mortgage lenders handle a consistent set of documents for every loan. These documents include earnings statements, W-2s, bank statements, driver’s licenses, voided checks, and insurance applications. Lenders must process these documents at scale, which involves classifying, extracting, and validating high volumes of data. The challenge is not unique to mortgage lending; organizations in banking, insurance, healthcare, and the public sector also face similar document processing issues, each with their own specific document types and compliance requirements.
To illustrate this, let's consider Summit Mortgage, a fictional mid-size lender that processes approximately 50,000 loans per year. Their loan processors spend 15-20 minutes per file manually sorting documents, verifying their completeness, and inputting data into the loan origination system. This manual process adds up to over 15,000 hours of manual document handling each year.
The complications multiply quickly. Manual data entry introduces errors that require additional work downstream. Incomplete documents are often left in queues until someone notices the missing information. During peak lending seasons, Summit had to rely on temporary staffing, which was expensive, slowed down when new staff were onboarded, and inconsistent in quality.
To address these issues, Summit Mortgage aimed to reduce document processing time from 15-20 minutes to under 6 minutes per file. They also sought to minimize data entry errors and manage peak volumes without hiring more staff. To achieve these goals, Summit deployed two AWS solutions: the AWS Generative AI Innovation Center Intelligent Document Processing (IDP) Accelerator and Amazon Quick Automate.
These tools were designed to automate the entire document intake pipeline, from the moment a lending package arrives to when validated data flows into the downstream systems.
The GAIIC IDP Accelerator is an open-source, serverless pipeline that processes documents at scale. Powered by Amazon Textract and Amazon Bedrock foundation models, the accelerator converts raw documents into machine-readable text, classifies each document, extracts structured data, and checks the extracted data against expected schemas.
Any anomalies, such as missing fields, inconsistent figures, or incomplete forms, are flagged for human review. The accelerator scales automatically with the volume of documents, charging only for the documents processed, with no need for infrastructure management.
Once the data is extracted, Quick Automate comes into play. It is an AI-powered automation service featuring a visual workflow builder and an AI assistant, allowing users to create multi-step workflows that include decisions, API calls, and system actions—all without custom coding. Summit configured their lending workflow in Quick Automate to route the extracted borrower data to the loan origination system, trigger income verification checks, flag incomplete packages with specific notifications for missing documents, and assign human review tasks for complex cases.
The end-to-end process works as follows: documents are received, the IDP Accelerator classifies and extracts the data, Quick Automate then orchestrates the downstream workflow, and finally, the validated data is used by loan processors, freeing them from tedious data entry tasks. This approach led to significant improvements in Summit Mortgage's operations:
- Document processing time was reduced from 15-20 minutes to under 6 minutes per file, achieving up to a 70 percent improvement.
- Summit was able to handle peak volumes without resorting to temporary staffing.
- Automated extraction and validation markedly reduced the number of manual keying errors that previously required rework.
In summary, by leveraging the AWS Generative AI Innovation Center IDP Accelerator and Amazon Quick Automate, Summit Mortgage successfully streamlined its document processing pipeline. The result was a substantial reduction in processing time, minimized errors, and increased efficiency, allowing the company to handle higher volumes of loans without the need for additional headcount.
This example demonstrates how scalable, serverless solutions can be tailored to the document-intensive operations of lenders and other industries, bringing significant business benefits.
Written by urgent.news from AWS Machine Learning's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
