Document Guide

Convert Closing Disclosure PDFs to Excel

Extract loan terms, itemized fees, and APR data from the CFPB three-page Closing Disclosure into clean Excel or CSV rows — no manual retyping of Sections A through H.

The CFPB Closing Disclosure (Form H-25) packs loan terms, projected payments, and itemized closing costs into a rigid three-page layout with lettered fee sections (A through H) and nested paid-by/paid-when columns. Manually transcribing even one disclosure into a spreadsheet means cross-referencing page 1's loan summary with page 2's fee tables and page 3's cash-to-close calculation. pdfexcel.ai uses AI field extraction to read digital and scanned Closing Disclosures and output structured rows — one per document — with fields like Loan Amount, APR, Total Closing Costs, and Cash to Close, ready for audit, comparison, or reporting.

Who This Is For

  • Mortgage brokers reconciling final terms against the Loan Estimate
  • Title and escrow processors building fee-comparison spreadsheets across multiple closings
  • Loan auditors and QC reviewers checking APR and Total Interest Percentage (TIP) consistency
  • Real estate accountants tracking closing costs across a portfolio of transactions
  • Compliance teams sampling disclosures for TRID timing and disclosure accuracy reviews

When This Is Relevant

  • You have a batch of Closing Disclosures from different lenders and need to compare Origination Charges (Section A) side by side
  • You're auditing a loan file and need APR, Total Interest Percentage, and Total Closing Costs pulled into one row per file
  • You're tracking Cash to Close and Loan Amount across a pipeline of closings for reporting
  • You need to convert dozens of PDF closing packages into a spreadsheet without retyping Sections A through H by hand

Supported Inputs

  • Digital PDF Closing Disclosures generated by loan origination systems
  • Scanned PDF copies from title companies or settlement agents
  • PNG or JPEG photos of printed Closing Disclosure pages
  • Multi-page PDF packages containing the three-page CD form

Expected Outputs

  • Excel (.xlsx) spreadsheet with one row per Closing Disclosure and columns for key fields (Loan Amount, Interest Rate, APR, Total Closing Costs, Cash to Close)
  • CSV export for loading into audit tools, Excel, or database imports

Common Challenges

  • Page 2's fee tables (Sections A-H) use nested sub-columns for Paid by Borrower, Paid by Seller, and Paid Before/At Closing — a straight table copy tool flattens these into one confusing column
  • Scanned or faxed Closing Disclosures from title companies often arrive slightly skewed, which throws off column alignment for dollar amounts during OCR
  • The loan calculations on page 3 (Total of Payments, Finance Charge, Amount Financed, APR, TIP) are formatted as a narrow table that's easy for rule-based parsers to misread
  • Redacted or blacked-out account numbers and borrower SSNs can cause adjacent fields in the same row to be dropped or blanked

How It Works

  1. Upload one or more Closing Disclosure PDFs, scans, or photos to pdfexcel.ai
  2. AI detects the CFPB three-page CD structure and identifies fields like Loan Term, Loan Amount, APR, and Total Closing Costs
  3. Select or customize which fields to extract — for example, just Section A Origination Charges and the Section J total, or the full loan calculation block from page 3
  4. Export the results as Excel or CSV, with one row per document, ready to compare, audit, or load into another system

Why PDFexcel.ai

  • Batch processing lets you run a folder of Closing Disclosures at once instead of opening each PDF individually
  • Custom field selection means you can target specific CD fields like APR, TIP, or Cash to Close rather than extracting the entire document
  • OCR handles scanned and photographed disclosures from title companies, not just clean digital PDFs
  • Pipeline automation with folder-based watch and export supports recurring workflows for lenders or auditors reviewing new closings on an ongoing basis

Limitations

  • The itemized fee tables in Sections A through H can include many line items with sub-columns; very dense or non-standard fee breakdowns may need manual review after extraction
  • Handwritten lender annotations or margin notes on a Closing Disclosure are recognized far less reliably than the typed form fields
  • Heavily redacted disclosures (blacked-out SSNs, account numbers) can result in missing or blank adjacent fields
  • State-specific or lender-customized CD variants that deviate from the standard CFPB layout may require custom field mapping rather than default extraction

Example Use Cases

  • A mortgage broker compiles 20 Closing Disclosures from different loan files into one spreadsheet to compare Origination Charges and APR across lenders
  • A title company processor extracts Cash to Close and Total Closing Costs from a batch of scanned CDs to reconcile against escrow ledgers
  • A loan auditor pulls APR and Total Interest Percentage from a sample of closed files to check TRID compliance thresholds
  • A real estate accountant builds a running spreadsheet of closing costs by pulling Section A-H totals from each new Closing Disclosure as loans close

Frequently Asked Questions

Can I extract just the APR and Total Closing Costs without pulling the entire Closing Disclosure?

Yes. The custom field selection feature lets you choose specific fields like APR, Total Interest Percentage (TIP), and Total Closing Costs (Section J) instead of extracting every field on the form.

Will this work on a scanned Closing Disclosure from a title company, not just a digital PDF?

Yes, OCR is built in for scanned PDFs and photos. Accuracy is best on documents with straight, well-lit pages — a skewed or angled scan can misalign fee columns on page 2, so straightening the scan first helps.

How does this handle the nested fee tables in Sections A through H on page 2?

AI-based extraction reads the paid-by and paid-when sub-columns in context rather than copying the table row by row, which is where simpler PDF table exporters (like a basic Acrobat export) tend to lose structure. Very dense, multi-tier fee breakdowns may still need a manual check afterward.

Can I process multiple Closing Disclosures at once and get one spreadsheet?

Yes, batch processing lets you upload multiple CD files and get a single spreadsheet with one row per document, which is useful for comparing loan terms or fees across several closings.

Ready to extract data from your PDFs?

Upload your first document and see structured results in seconds. Free to start — no setup required.

Get Started Free

Related Resources