PDF to CSV: The Format Your Import Job Actually Accepts
Excel is where humans look at data. CSV is where software reads it. The two require different discipline: a CSV with commas inside unquoted descriptions, or with thousand separators still attached, will silently corrupt an import.
Convert one file free — see the table land in Excel
Free daily conversions · CSV, TSV, and Excel export · files shredded after processing.
Why CSV exports fail downstream
Commas inside descriptions
A merchant named "Acme, Inc." splits one row into two columns unless the field is quoted. Every row after it shifts.
Thousands separators
1,204.55 in an unquoted CSV is two fields. Imports either reject the row or store 1 and 204.55 separately.
Encoding surprises
Accented characters turn into mojibake when the file is read as Latin-1 instead of UTF-8.
Date formats that guess
03/04/2026 is March 4 or April 3 depending on locale. An import does not ask.
Header rows repeated mid-file
Copied PDFs insert the page header every 50 rows; the import treats each one as a transaction.
Empty cells collapsed
Some converters drop trailing empty fields, so rows have different column counts.
CSV hygiene checklist
- Every text field quoted, so commas and quotes inside descriptions are harmless.
- Numbers exported without currency symbols or thousands separators; keep the currency in its own column.
- One date format across the file, ideally ISO 8601 (2026-03-04).
- A stable column count on every row, including empty trailing fields.
- UTF-8 encoding, with a BOM only if the target tool needs it (Excel on Windows often does).
Importing into bookkeeping software
QuickBooks, Xero, and most bank feeds want a signed amount column or separate debit/credit columns, plus a date and a description. They do not want subtotal rows, page headers, or a summary block at the end.
That means the conversion should already separate transactions from summaries. If your CSV contains a "Total debits" row, the import will either fail or create a phantom transaction.
Using CSV in a pipeline
If you are scripting, export CSV and treat the file as immutable input: read it, validate row count and column count, then write to your database. Never edit the CSV by hand between conversion and load — that is where reproducibility dies.
For a document set, keep the original PDF filename in a column. It is the only reliable way to trace a bad row back to the page it came from.
Frequently asked questions
What is the difference between CSV and TSV export?+
CSV separates fields with commas and needs quoting for text; TSV uses tabs and is more tolerant of commas in descriptions. Paste-from-clipboard workflows use TSV because Excel splits it on paste automatically.
Will the numbers import as numbers?+
They will if the export strips currency symbols and separators. Keep the currency as a separate column to avoid ambiguity.
Can I convert a scanned PDF to CSV?+
Yes — OCR runs first, then the table is structured and exported. Expect to check a sample of rows for misread digits.
Is CSV better than Excel for importing?+
Usually. XLSX carries formatting and multiple sheets; CSV is simpler and more predictable for import jobs. Use XLSX when a human will be editing the file.
Does it work for bank statements?+
Yes, and for statements the CSV is normally used for a bookkeeping import. The statement-specific path also verifies the running balance before export.
Keep reading
Done reading? Convert a document and check the output yourself.
Convert to CSV free