LOCAL FILE PROCESSING
Extract a PDF table to Excel without uploading the PDF
For a PDF whose table cells contain selectable text, PdfTableKit reads the file and creates the spreadsheet in your browser. Choose one page or table region, correct the grid and download XLSX. The selected PDF and grid are not sent to our billing service.
What stays local, and what still uses the network
The PDF bytes, extracted text, source preview and edited grid are processed in the browser tab. The website and its pdf.js and SheetJS libraries still require network requests. The hosting service and library CDN receive ordinary request information. This is local document processing, not an offline application or a promise that visiting the website makes no network requests.
Check the table cells before choosing the file
Open the PDF in a reader and try selecting the actual words and numbers inside the table. Being able to select the report title does not prove the table has a text layer. A scanned table can sit below a selectable caption. Image tables, scans and screenshots require OCR, which this tool does not provide. Password-protected PDFs are also outside this version.
Try a public synthetic example first
Download our two-page fictional sample and open page 1 in the extractor. The small table should include Apples, quantity 2 and price 1.50. This is a reproducible example, not customer data or an accuracy benchmark. Use it to check the upload, editing and XLSX download workflow before choosing your own authorized file.
Review a selected page or region
Files are limited to 12 MB. One Extract processes the selected page; the free allowance is 3 text pages per day (resets 00:00 UTC). Draw a region on the source preview to exclude surrounding notes, then compare the grid with the complete printed table. A region is a selection aid, not a guarantee that all rows or column boundaries are correct.
Export a checked spreadsheet
Correct split labels, missing rows and misaligned columns before downloading. XLSX cells retain the grid as text, including leading zeros and formula-like strings. Amounts are not automatically ready for arithmetic; deliberately convert the appropriate columns in your spreadsheet app after proofreading.
When to choose another workflow
If the table is an image, obtain an authorized text-based export or use a separate OCR workflow and check its output. For complex merged headers, rotated text or a table spanning multiple pages, do not assume this single-page tool reconstructs the whole document. Stop if the grid cannot be reconciled with the source.
Questions
Do I need an account for the free extractor?
No. The 3-page daily free allowance is counted on our server with a functional cookie, no account needed. Paid access, when available, uses a private purchase key rather than Google sign-in.
Does local processing mean the website works offline?
No. Website assets and external libraries need to load. Local processing refers to your selected PDF and extracted table, not all network activity.
Does the payment provider receive my PDF?
No PDF or grid content is part of the billing requests. Payment processing uses purchase and order information separately.
Can I extract a scanned bank statement?
Image-only and scanned tables are not supported. Even selectable-text financial tables require manual checking; this is not an accounting or audit service.