The Criners LLC · Powered by Money Hunter
From a public digital PDF to usable text
This sample shows shape: what we could read, what we could not, and the extracted table. The rows below were invented for this demonstration.
Executive summary
The source is a short digital PDF (selectable text), not a scan. Two tables extracted. Page 3 is an image of a stamped form and produced no reliable rows — that limit is written below, not guessed.
Findings
- Pages 1–2 contain selectable text and a two-column price table.
- Headers used mixed units (“Qty ” with a trailing space).
- Page 3 is an image-only stamp; no OCR was run on this sample.
Evidence — as read from the PDF
| Item | Qty | Unit |
|---|---|---|
| Widget A | 12 | box |
| Widget B | 4 | box |
Recommendations
- Use the extracted CSV for imports; do not re-type from the PDF.
- If later pages are scans, send a digital export instead of a photograph.
- Cleanup of the resulting CSV is a separate $49 job if headers still need snake_case.
Extracted table
| item | qty | unit |
|---|---|---|
| Widget A | 12 | box |
| Widget B | 4 | box |
Next steps
A paid job needs a public PDF URL in the Stripe notes. Scanned pages are limited by file quality.