Bulk OCR now handles tables as well as plain text. Upload a scanned invoice, receipt, or bank statement, and FileDrop returns an Excel file with real rows and columns, ready to sort, sum, and filter.
This replaces the separate PDF to Excel tool; /pdf-xlsx now opens Bulk OCR with the Table or spreadsheet document type selected.
Pick the kind of document
Instead of choosing an OCR engine yourself, start with a document type. Each one presets the engine, table mode, and output format:
| Document type | Best for | Default output |
|---|---|---|
| Text document | Letters, contracts, articles, scanned pages | Text |
| Table or spreadsheet | Data laid out in rows and columns | Excel |
| Invoice | Line items, quantities, totals | Excel |
| Receipt | Purchased items and amounts | Excel |
| Bank statement | Transactions with dates and amounts | Excel |
| Complex table (high accuracy) | Borderless or irregular tables (slower) | Excel |
| Handwriting | Handwritten notes and forms (slower) | Text |
You can still change the engine, language, or output afterwards. Doing so switches the job to a custom setup.
Tables keep their layout
Table jobs rebuild rows and columns from the position of each word on the page. Empty cells stay empty, and right-aligned numbers stay in the right column. A table that runs across several pages is stitched into one, with repeated headers and titles removed.
In the Excel file, each source file gets its own sheet. Numbers, currency amounts, percentages, and unambiguous dates become real values with number formats, so totals work straight away. IDs, phone numbers, and ambiguous dates stay as text so nothing is changed by mistake. Columns are auto-sized. Word exports contain the same tables.
Open results in Google Sheets or Docs
When a job completes, choose Open in Google Sheets or Open in Google Docs to continue in a native Google file, with no download and re-upload.
Other improvements
- One file is enough. Bulk OCR no longer needs a minimum of two files.
- Longer PDFs. PDFs are read in 3-page parts and joined back together, up to 30 pages.
- Language auto-detect is now the default.
- More reliable jobs. Redelivered jobs no longer show a false Failed status, and a server error on Engine 3 falls back to Engine 2.
- Available on the Google Add-on plan as well as other paid plans.
Example: Upload a month of scanned supplier invoices, choose Invoice, and start the batch. Download one Excel file with a sheet per invoice, or open it directly in Google Sheets.
The Bottom Line:
One keeps you awake. The other gets work done.
A month of coffee: $80
A month of FileDrop: $19
Why not have both?
