How it works
You choose the structure. AI handles the reading.
The workflow is designed for repeated documents: define the information once, apply it to every PDF, then review one organized table.
- 01
Upload your PDFs
Select invoices, receipts, or similar PDF documents. Each file is treated as a separate record.
- 02
Define the Excel columns
Name the fields you need and optionally describe what each field means or choose its data type.
- 03
Extract each document
The AI reads one PDF at a time and returns only values that fit your requested structure.
- 04
Review the preview
Every PDF appears as one row. Values that cannot be located confidently remain blank.
- 05
Download the workbook
Completed rows are combined into one .xlsx file using your column names.
A concrete example
Three PDFs become three rows.
Suppose you upload three invoices and request Company, Invoice Number, Date, and Amount. ExtractToExcel applies the same schema to every file and preserves the source-to-row relationship.
| Source PDF | Company | Invoice no. | Date | Amount |
|---|---|---|---|---|
| invoice-001.pdf | ABC Sdn Bhd | INV-001 | 2026-09-01 | 450.00 |
| invoice-002.pdf | XYZ Ltd | INV-839 | 2026-08-24 | 1,220.50 |
| receipt-003.pdf | Cafe One | 2026-09-03 | 29.90 |
What happens when something is missing?
The corresponding cell stays empty. ExtractToExcel is designed to avoid guessing a value just to complete a row. You should still review important financial and business data before relying on the workbook.