Batch OCR that turns 100 files into one spreadsheet.
Pick a Parser Template, drop in up to 100 PDFs, scans or photos, and press Extract Data. Every document comes back as rows in one sheet, ready for Excel, CSV or Google Sheets.
Convert multiple PDFs to Excel, even when they all look different.
Different customers, different layouts, different file types. One Parser Template reads them all, so a PDF from one customer, a scanned order from another and an email from a third land in the same columns.
Bulk OCR in three steps, no code.
All you need is a Parser Template. Build one from a sample with the AI document parser, or pick a ready-made one.
Pick a Parser Template
Choose the template for this document type. If a Google Sheet or Airtable table is linked to it, Data Lab tells you the rows will go there too.
Drop in your files
Up to 100 files, 50 MB each. Before you start, you see how many page credits the batch will use and how many you have left.
Extract, check, export
Press Extract Data and watch each file finish. Check the rows in the sheet, then export them in one click.
One batch, four ways out.
Every export has the same columns: one per field, a Source File column, and a row for each line item. Click a format to see it.
An .xlsx file with one column per field. Line items get one row each, so the file is ready for a pivot table or a VLOOKUP.
| A | B | C | D | E | F | G | |
|---|---|---|---|---|---|---|---|
| 1 | Source File | order_number | order_date | buyer_name | total_amount | line_items.description | line_items.quantity |
| 2 | acme-retail_po-55102.pdf | PO-55102 | 2026-10-12 | Acme Retail Ltd | 1386.00 | Kraft mailer box, small | 1200 |
| 3 | Kraft mailer box, large | 600 | |||||
| 4 | greenleaf_order-7781.pdf | 7781 | 2026-10-13 | Greenleaf Foods | 2104.50 | Insulated liner, 30 cm | 450 |
| 5 | polar-outdoor_scan.jpg | PO/2026/339 | 2026-10-13 | Polar Outdoor Co. | 418.00 | Poly mailer, grey | 1900 |
| 6 | meridian_PO88213.xlsx | 88213 | 2026-10-14 | Meridian Health Supplies | 960.00 | Tamper-proof seal | 8000 |
Skip the export: rows land in your sheet on their own.
Link a Google Sheet or an Airtable table to the Parser Template once. From then on, every document you extract in Data Lab is added to it automatically. Columns are made for every field, and each line item gets its own row.
Set-up guides: PDF to Google Sheets and Airtable OCR.
Need last month's batch again? It is in Activity.
Left the page before you exported? No problem. Every Data Lab batch is listed in Activity. Open it to see the same spreadsheet and export it again in any format.
- Filter by Parser Template and date
- See how many documents succeeded or failed
- Export again as Excel, CSV, PDF or Google Sheets
How many page credits will your batches use?
One page credit per page, whatever the format. A photo of a receipt is one credit; a 3-page PDF is three.
Data Lab, automations or the API?
All three use the same Parser Template, so you can start in Data Lab and automate later. See how ParserBee works for the full picture.
Data Lab
For the files you have now: month-end invoices, a box of scanned orders, a year of bank statements. Upload, extract, export.
How it works ↑Automations
For documents that keep arriving. Zapier, Make or n8n send each new attachment or Drive file to ParserBee with no upload.
ParserBee integrations →API
For your own software. Send a file and a Template ID, get the same fields back as JSON.
Document parsing API →Data Lab questions
What is batch OCR?
Batch OCR reads many documents in one go instead of one at a time. Data Lab goes a step further than plain OCR: instead of raw text, every document comes back as named fields (an order number, a date, a total, each line item) in one spreadsheet.
How many files can I upload at once?
Up to 100 files in one batch on a paid plan, and 10 on the free trial. Each file can be up to 50 MB, and PDFs up to 24 pages (12 on the free trial). For more files, run another batch.
Which file types does Data Lab accept?
PDF, JPG, PNG, WebP and HEIC, plus Excel (XLSX), CSV, PowerPoint, email (EML), plain text, RTF, HTML, ODT and older Word (.doc) files. DOCX and the old XLS format are not accepted: save them as PDF first.
Can one batch hold documents with different layouts?
Yes. Mix freely: documents from different companies, in different designs, as PDFs, scans or photos, can all go in one batch. Use one batch per document type, because each batch runs with one Parser Template.
How do I convert multiple PDFs to Excel?
Pick a Parser Template in Data Lab, drop in your PDFs, press Extract Data, then click Download as Excel. You get one .xlsx file with a column per field, a Source File column, and a row for each line item.
How are line items laid out in the export?
Each line item gets its own row. A document's own fields, such as the order number and total, are on its first row, and the Source File column tells you which file each row came from.
What happens if a file fails?
The rest of the batch carries on. When the batch is done, the Batch Complete card shows how many succeeded and failed, and lists each failed document with the reason.
How many page credits does a batch use?
One page credit per page. Before you start, Data Lab shows how many page credits the batch will use at least and how many you have left. New accounts get 25 free pages; paid plans start at $29 a month for 250 pages. See pricing.
Can I export a batch again later?
Yes. Every batch is listed in Activity. Open it to see the same spreadsheet and export it again as Excel, CSV, PDF or Google Sheets.
Can Data Lab send the rows to Google Sheets or Airtable automatically?
Yes. Link a Google Sheet or an Airtable table to the Parser Template once, and every document you extract from then on is added to it. Or use Export to Google Sheets to make a new spreadsheet from one batch.
Can Data Lab pick up files on its own?
Is my data used to train AI?
No. Your documents are never used to train AI models, and files travel to ParserBee over an encrypted (HTTPS) connection.