← Back to app

About Statement Parser

A tool for converting Japanese property management PDF statements into structured CSV data.

What it does

Property management companies in Japan issue monthly settlement statements (月次精算書) as PDFs. These are detailed but not machine-readable — getting the numbers into a spreadsheet means manual data entry.

This tool reads those PDFs using AI and extracts the structured data automatically: room-by-room income, expense categories, and monthly totals. The output is clean CSV files you can open directly in Excel.

How it works

1
Upload PDFs

Drag and drop one or more monthly statement PDFs. Up to 15 at a time.

2
AI extraction

Each PDF is sent to an AI model (Google Gemini by default) which reads the statement and returns structured data — income per room, expenses by category, and the net transfer amount.

3
Review and download

Results appear immediately. Download individual monthly CSVs, or click "Download ZIP" to get all months plus an aggregated annual pivot in one file.

What's included in the output

What it's designed for & what it's not

✓ Designed for

  • Japanese monthly property management statements (月次精算書)
  • Getting PDF data into Excel quickly
  • Comparing income and expenses across multiple months
  • Processing statements from multiple buildings

✗ Not designed for

  • Accounting, bookkeeping, or tax filing
  • Managing tenants, leases, or contracts
  • Non-Japanese statement formats
  • 100% guaranteed accuracy — AI can misread unusual layouts
  • Storing or archiving your documents
Always verify the output. AI extraction is accurate for standard statement formats but should be spot-checked against the original PDF, especially for statements with unusual layouts or handwritten sections.
Don't close your browser before downloading. Parsed results are held in your browser tab only — they are not saved to the server. If you close or refresh the page before downloading, the results will be lost and you will need to parse the files again.

Data and privacy

PDFs are deleted from the server immediately after processing. Parsed results exist in your browser session only and are never stored on the server. See the Privacy Policy for full details.