Extract Comments and Annotations from PDF
Export every comment, highlight and sticky note in a PDF to CSV, Markdown or JSON, with the page, author, date and highlighted text.
Select a PDF fileor drop it anywhere on this page, or paste
Opens in your browser. Nothing is uploaded.
Extract Comments and Annotations from PDF, free and private
Extract comments from a PDF and get every sticky note, highlight, text box and reply as one list. The tool reads the comment layer of the file and writes a row per comment with its page, type, author, date and text. For highlights, underlines and strikethroughs it also recovers the words under the mark, which the PDF itself does not store.
Save the list as CSV for Excel or Google Sheets, as Markdown for notes and tickets, or as JSON for scripts. The PDF is read in your browser on a Mac, a Windows PC, an iPhone or an Android phone. It is never uploaded, and you don't need Adobe Acrobat.
How to extract comments from a PDF
- 1
Add your PDF
Press Select a PDF file or drop the PDF on the page. The tool reads one document at a time and lists its comments by page as soon as the file opens.
- 2
Check the comment list
Each comment shows its number, type, author, date, the marked text and the note. Replies sit indented under the comment they answer, and a colour dot shows the colour of the mark. Untick a type or an author to leave those comments out. The list shows the first 200 comments, and the export has every comment that passes the filters.
- 3
Pick CSV, Markdown or JSON
CSV gives one row per comment for a spreadsheet. Markdown gives a readable report with a heading per page. JSON gives the same fields as structured data. A preview of the file appears beside the list, with a Copy button.
- 4
Export the PDF comments
Press Export comments, then press Download. The file is named after your PDF with -comments added, such as contract-comments.csv.
Why use PDFToolz for this
The words under each highlight
A PDF stores a highlight as a shaded area, not as text. The tool matches that area against the words on the page and fills a Marked text column. A highlight with no note still tells you what was marked. Underlines, strikethroughs and squiggly underlines work the same way.
Replies and review status
Each reply follows the comment it answers and gives that comment's number in the Reply to column. When a reviewer sets a status in Adobe Acrobat, such as Accepted or Rejected, the latest one goes in the Status column instead of a row of its own. A status set back to None leaves the column empty.
Every kind of comment
Sticky notes, text boxes, inserted text marks, drawings, rectangles, ovals, lines, polygons, stamps, file attachments and redaction marks are all listed, each with a plain type name. Links and form fields are not comments, so they are left out.
CSV that opens cleanly in Excel
The CSV is UTF-8 with a byte order mark, so accents and non-Latin scripts show correctly in Excel. A comment that starts like a formula, such as =SUM(A1) or @name, gets a leading apostrophe, so Excel shows it as text and does not run it as a formula.
What each export contains
All three formats hold the same fields. Number is the position in the list. Page is the page the comment sits on. Type is a plain name such as Note, Highlight or Drawing. Author is the name the PDF app saved with the comment.
Date is when the comment was last changed, or when it was made if the file has no change date. It is shown as the file wrote it, such as 2024-03-01 09:30, without converting time zones. Color is the comment color as a hex code such as #ffff00, which helps when reviewers used colors to mean different things.
- CSV has the columns Number, Page, Type, Author, Date, Comment, Marked text, Reply to, Status and Color. Line breaks inside a comment stay inside its cell.
- Markdown has a heading per page and a numbered subheading per comment. The marked text is a quote and the note follows it.
- JSON is one object with the source file name, the count and a comments array. Empty fields are null, so every entry has the same keys.
How to export PDF comments to Excel or Google Sheets
Choose CSV and export. Excel opens the file directly with a double click. In Google Sheets, open the File menu, choose Import and upload the CSV.
Sort by Page to follow the document, or by Author to split the work between reviewers. Filter the Status column to find comments nobody has accepted or rejected yet.
How the highlighted text is found
Each highlight stores the corners of the area it covers. The tool places every character of the page text along its line, keeps the characters inside that area and rounds the result to whole words. A word counts when at least half of its letters are inside the mark.
Character widths are estimated from the letters, not read from the font. A highlight that stops partway through a word therefore returns the whole word. A scanned page is a picture with no text layer, so the Marked text column stays empty there while the note itself is still exported.
Extract comments without Acrobat, on Mac, Windows or iPhone
The tool runs in a web browser, so it works the same in Safari on a Mac or iPhone, in Chrome or Edge on Windows, and on Android. There is nothing to install and no account to create.
Long review files are read one page at a time, and the page text is read only on pages that carry highlights or other text markup.
What the tool can't extract
Only comments stored as PDF annotations are found. Marks that were printed into the page, flattened by another program, or written on paper and scanned are part of the page, so they can't be listed. If the tool reports no comments, that is usually why.
Hidden comments are included. Annotations that Acrobat groups into one comment are listed once. The personal check marks Acrobat lets each reviewer set are left out. Files that open without a password but block copying or editing work as normal. If the PDF asks for a password to open, remove it with Unlock PDF first.
Frequently asked questions
Add the PDF, pick CSV, Markdown or JSON and press Export comments. Every note, highlight and reply is saved with its page, author and date. It is free, and the file is not uploaded.
Yes. Choose CSV. The file opens in Excel with one row per comment and columns for page, type, author, date, comment, marked text, reply and status.
Yes, when the page has real text. The Marked text column holds the words under each highlight, underline or strikethrough, rounded to whole words. Scanned pages have no text layer, so that column stays empty for them.
Choose Markdown. The result is a plain text file with a heading per page. Any text editor opens it, and apps such as Obsidian, Notion or GitHub show it as a formatted report.
Yes. Replies follow the comment they answer and give its number in the Reply to column. Statuses set in Acrobat, such as Accepted or Rejected, appear in the Status column.
Yes. The tool runs in a web browser on Windows, Mac, iPhone, iPad and Android. You don't install anything or sign up.
The marks are not annotations. A PDF that was printed, flattened or scanned has its comments drawn into the page, and those can't be read as separate comments. Links and form fields don't count as comments either.
No. The PDF is read in your browser and the export is built on your device. Nothing is sent to a server.
