Cync Software
Automating Financial Statement Processing for Faster Credit Decisions
Minutes
End-to-End Processing
80% Reduction
In Mapping & Extraction Time
GAAP Mapping
Automated via NLP
About
A lending software product company focused on creating software solutions to manage Asset-Based Lending (ABL) loans for commercial finance companies and banks.
Industry
Financial Technology (Lending & Credit Management Software)
The Challenge
Cync Software and its lending clients faced a significant bottleneck in processing borrower financial statements for credit analysis.
For banks and financial lenders, assessing credit risk requires obtaining and analyzing borrower financial statements including Balance Sheets, Income Statements, and Cash Flow Statements — to calculate key financial ratios. Manually entering this data into spreadsheets or third-party applications was an extremely time-consuming and error-prone process.
Adding to this complexity, Small and Medium Enterprises frequently use a wide variety of unique financial terminologies in their statements. Mapping these non-standard terms to GAAP-compliant categories required Subject Matter Expertise (SME) and any errors in this mapping directly impacted final credit ratios and lending decisions.
The client's existing system also lacked the capability to process scanned PDF documents, forcing lenders to manually re-enter financial values before any analysis could begin.
The Solution
Using InferIQ's financial statement processing platform, Cync automated the extraction, mapping, and analysis of financial statement data across multiple document formats.
1. Extracting financial content using AWS Textract
InferIQ leveraged AWS Textract to extract content from both live-text and scanned PDF documents. Unlike traditional OCR solutions, AWS Textract delivered a significantly higher level of accuracy enabling reliable extraction even from low-quality scanned files.
2. Custom ML model for GAAP mapping
InferIQ's custom-built ML model leveraged Natural Language Processing (NLP) to accurately align and map each borrower's unique financial terminology to GAAP-standard financial terms. Once mapped, the platform automatically validated the data and auto-calculated the required credit ratios for underwriting analysis.
3. A Human-in-the-Loop feedback mechanism
Allowed analysts to review and correct mappings when necessary continuously retraining the model and improving its accuracy over time with every interaction.
The Impact
By automating financial statement spreading, Cync dramatically reduced the time and effort required for credit analysis.
What previously took days of manual work mapping non-standard financial terms, re-entering data, and calculating ratios by hand is now completed in minutes. The solution also introduced scanned document processing for the first time, expanding the range of financial statements the platform could handle.
With financial ratios calculated automatically and GAAP mapping handled by the ML model, lending teams were freed from repetitive data entry and able to redirect their focus toward core business priorities and customer-facing activities.
Key Results
Processing time for mapping and extracting financial data reduced by 80%
Financial statement analysis accelerated from days to minutes
GAAP-compliant financial term mapping fully automated via NLP
Scanned document processing introduced for the first time
Human-in-the-Loop feedback continuously improves model accuracy
Lending teams able to focus on higher-value credit analysis work