How to Convert a PDF to CSV on Mac: Free Local Methods That Do Not Upload Your File
CSV is the plain, universal format for table data, which is why accounting software, databases and analytics tools ask for it. Getting your data out of a PDF and into one is a common chore. You can do it on a Mac without uploading the file to a website, and for many PDFs the result is cleaner than what an online converter produces.
Quick answer: If the PDF has selectable text, copy the table into Excel, clean it, then choose File > Save As, set the format to CSV UTF-8 (Comma delimited) and save. For many tables or an awkward layout, use the free, open-source Tabula app, which runs on your Mac, lets you draw a box around each table and exports to CSV. Neither works on scanned pages without text recognition first.
Check the PDF type first
Everything here depends on whether the PDF contains text or only a picture of text. Open it in Preview and try to select a word in the table, or search for one with ⌘F. If it highlights, the PDF has a text layer and the methods below work. If not, it is a scan and you need OCR before any of them can extract anything.
| Method | Cost | File leaves your Mac? | Works on scanned PDFs? | Effort | Best for |
|---|---|---|---|---|---|
| Copy into Excel, Save As CSV UTF-8 | Excel licence | No | No | Low | One or two tables |
| Tabula | Free, open source | No (runs locally in your browser) | No, text-based PDFs only | Low to medium | Many tables, awkward layouts |
| pdftotext -layout, then import | Free (Homebrew install) | No | No | Medium | People comfortable with Terminal |
| Online PDF to CSV converter | Free tiers, limits apply | Yes | Sometimes, with OCR | Low | Non-sensitive files only |
Method 1: Copy into Excel, then save as CSV
For one or two tables this is the quickest route, and it gives you a chance to clean the data on the way.
Copy the table from the PDF and paste it into Excel. Our guide to copying a PDF table into Excel on Mac shows how to select cleanly.
Split and clean the data, following the guide to cleaning pasted PDF data. Make sure numbers are numbers and dates are dates.
Press ⇧⌘S and choose CSV UTF-8 (Comma delimited) from the File Format menu.
Save, and accept the warning that only the active sheet is saved.
The UTF-8 option matters. The older plain CSV and Macintosh CSV options can turn accented characters and currency symbols into unreadable characters. Our guide to the Save and Save As shortcuts explains the CSV formats and the two limits to remember: CSV saves only the active sheet, and it stores the results of formulas, not the formulas.
Method 2: Tabula
Tabula is a free, open-source tool built for exactly this job, originally for journalists extracting tables from reports. It runs on your own Mac, and you use it in your browser at a local address, so the PDF is not sent anywhere. The project states that it works only on text-based PDFs, not scanned documents.
Download Tabula from the project's page and open it. Check the releases page for the current version and its requirements, because the project has not changed much in recent years and older notes mention Java.
If macOS blocks the app the first time, open System Settings > Privacy & Security and allow it.
Your browser opens to a local address. Upload the PDF.
Draw a box around each table you want. Tabula can also try to detect tables for you.
Preview the extraction, then export as CSV.
Tabula shines when the table has no visible column lines, because it looks at the spacing of the text and offers two extraction modes, one for tables with ruling lines and one for tables without. Try both on a stubborn table. It is also much faster than copy and paste when you have forty pages of the same layout.
Method 3: pdftotext in Terminal
If you are comfortable with Terminal, the Poppler tools can extract text with the layout preserved. Install them with Homebrew (brew install poppler), then run pdftotext with the layout option on your PDF. The output is a text file in which columns line up with spaces. You can open it in Excel with the Fixed width option in Text to Columns, as described in the cleaning guide. This is not a friendly route, but it is scriptable, which helps if the same report arrives every month.
Method 4: Online converters, with care
Online PDF to CSV converters are quick, and some include OCR. The cost is that your file is uploaded. For public documents that does not matter. For bank statements, payroll and customer data, use a local method. Check the retention policy, file size limits and whether the free tier adds anything to your output.
CSV traps to avoid
A CSV looks simple, but several things go wrong when data comes from a PDF.
Commas inside values. An amount such as 1,234.50 contains a comma. A proper CSV wraps it in quotes. Excel does this for you. Hand-built files often do not, and the amount splits across two columns. Remove thousands separators before you export if the destination cannot cope.
Encoding. Use UTF-8 so names and symbols survive.
Date formats. Some systems want 2026-03-04, others 04/03/2026. Check what the destination expects.
Leading zeros. Format codes as Text before saving, or Excel will have dropped the zeros already. Our guide to cleaning a CSV covers these problems.
Header row. Most importers expect exactly one header row with specific names. Check the importer's documentation, because accounting packages are strict about column names and order.
Multiple sheets. CSV holds one sheet. Save each sheet separately.
Check the result
Open the finished CSV in a plain text editor such as TextEdit set to plain text, and look at the first few lines. You should see one header line and one row per record, with values separated by commas and quotes only around values that contain commas. Then compare the row count and the total of the amount column against the PDF.
Common mistakes
Skipping the check for selectable text, and trying Tabula on a scan.
Saving as plain CSV rather than CSV UTF-8, then finding garbled characters.
Exporting with numbers as text, which some importers reject.
Not removing page headers and footers that were copied with the table.
Assuming one extraction mode suits every table. Switch modes and compare.
Frequently asked questions
What is the best free way to convert PDF to CSV on a Mac?
For a handful of tables, copy into Excel and save as CSV UTF-8. For many tables, Tabula is free and keeps everything local.
Can Numbers convert a PDF to CSV?
Numbers does not open PDFs as tables. You would copy the text and paste it into Numbers, then export as CSV, which is the same idea as using Excel.
Will converting keep formatting?
No. CSV holds only values, so there are no colours, fonts or formulas.
Does it work for bank statements?
It works if the statement PDF has selectable text. Check each row and the totals, and see the importer's requirements for the column layout it expects.
Is Tabula safe to use?
It is open source and processes files on your own computer. As with any download, get it from the project's official page and check what version you are installing.