Convert a PDF to Rich Text Format (RTF)
Converting PDF to RTF turns a fixed-layout PDF into a Rich Text Format document — an editable, plain-text markup format from 1987 that nearly every word processor ever made can open, including WordPad, TextEdit, Word, and LibreOffice. RTF keeps character formatting like bold, italics, and font sizes but has a far simpler layout model than PDF, so pages become flowing text. The dependable routes are Word's built-in PDF converter saved out as .rtf, a LibreOffice rebuild, or Calibre's command-line converter, all covered below.
Producing useful RTF from a PDF requires reconstructing paragraphs, styles, and tables from positioned glyphs — a layout-analysis engine that browsers do not provide, and that a client-side script cannot fake honestly. The desktop methods below use real converters and tell you exactly what each one loses.
Disclosure: this page contains affiliate placeholder links. If they were live, we could earn a commission at no extra cost to you.
AFFILIATE PLACEHOLDER: Recommended desktop PDF suite (placeholder link)
How to convert a PDF to RTF (step by step)
Method 1: Microsoft Word (best structure recovery)
- Open the PDF in Word. Choose File → Open and select the PDF. Word warns that it will convert the PDF into an editable document; click OK and wait while its reflow engine rebuilds paragraphs and tables.
- Review the converted document. Check headings, tables, and lists against the original PDF, fixing any misdetected paragraphs while the document is still in Word's richer format.
- Save as Rich Text Format. Choose File → Save As, set the file type to Rich Text Format (*.rtf), and save.
Method 2: LibreOffice (free)
- Open the PDF via File → Open; it loads in LibreOffice Draw as one frame per text block.
- Copy the text page by page (Ctrl+A, Ctrl+C) and paste into Writer with Ctrl+Shift+V → Formatted text (RTF), which keeps bold and font sizes.
- Save via File → Save As → Rich Text (.rtf).
Method 3: Calibre's ebook-convert (command line, batch-friendly)
Calibre installs a converter that writes RTF directly: ebook-convert input.pdf output.rtf. It performs its own paragraph reconstruction (including un-hyphenating line-wrapped words) and is scriptable over hundreds of files, though tables come out as plain lines and headers/footers repeat in the text. It only reads selectable text — scans yield nothing.
How it works
RTF is worth understanding because it explains both its universality and its limits. An RTF file is 7-bit ASCII markup: {\rtf1\ansi {\b Total due:} 4,182.37} renders "Total due:" in bold. There is no container, no compression, no schema — just nested groups and control words — which is why a 1994 word processor and Windows Notepad-era WordPad can both parse a file written today. Formatting is limited to what those control words cover: character properties, paragraph alignment and indents, simple tables (\trowd…\cell), and page margins. PDF concepts like layers, transparency, and absolute-positioned frames have no RTF equivalent, so converters must discard or approximate them.
The conversion itself is the standard structure-recovery problem. A PDF stores each visual line as positioned glyph runs with no paragraph markers; the converter sorts runs into lines, merges lines into paragraphs when leading and margins are consistent, detects headings by size jumps, and maps recovered runs to RTF control words — an 11pt bold run becomes \b\fs22 (RTF measures font size in half-points). Word's engine additionally detects table grids and list bullets, which is why its RTF output has real table rows where simpler tools emit tab-separated text.
Images are the format's weak spot: RTF embeds them as hexadecimal text under \pict, roughly doubling every image's size and bloating files fast. And as with every PDF export path, only selectable text can be recovered — a scanned PDF must go through OCR before any of these methods produce more than blank pages.
Worked example: a 6-page letter for a WordPad user
A reader needed a 6-page, 380 KB PDF letter (letterhead logo, bold headings, one small table) editable by a relative whose PC has only WordPad. Word 365 opened and converted it in 9 seconds, recovering all 41 paragraphs, the table, and the logo; saved as RTF the file was 1.9 MB — five times the PDF, almost entirely the hex-encoded logo. The same document saved as .docx was 210 KB, confirming the compression difference. WordPad opened the RTF flawlessly except the table borders, which WordPad's simpler renderer drew thinner. The LibreOffice route on the same file took 12 minutes of reflowing and lost the table grid; Calibre's ebook-convert took 20 seconds but dropped the logo and flattened the table to text lines.
Frequently asked questions
Why choose RTF instead of DOCX when converting a PDF?
RTF opens in virtually everything, including WordPad on Windows, TextEdit on macOS, and decades-old word processors, with no ZIP container or XML schema involved. Choose it when the recipient's software is unknown or old; choose DOCX when you need styles, comments, or modern layout features.
What formatting survives a PDF to RTF conversion?
Character formatting such as bold, italics, font size, and color survives well, and simple tables usually make it through the Word route. Columns, text boxes, headers and footers, and precise page geometry frequently do not.
Can I convert PDF to RTF without Microsoft Word?
Yes — open the PDF in LibreOffice Draw, copy the text into Writer, and save as .rtf, or use Calibre's ebook-convert command line which writes RTF directly from a PDF. Both are free; both need more manual reflow work than Word's converter.
Do images end up inside the RTF file?
They can — RTF embeds pictures as hex-encoded data in the text stream, which is why an RTF with images is often several times larger than the same DOCX. Word's conversion carries images across; plain-text routes drop them.
Why is my converted RTF file so large?
RTF is an uncompressed plain-text format, and every embedded image is stored as hexadecimal text, roughly doubling its byte size. If size matters, save as .docx instead, which ZIP-compresses everything.