How to Edit Text in a PDF for Free (and Why Most Editors Charge for It)

Why a PDF is not a Word document

A Word file stores paragraphs: the program decides where each line breaks every time you open it. A PDF stores the finished page. Inside each page is a content stream, a list of drawing instructions that say, in effect, “select this font, move to this point, paint these glyphs”. There is no paragraph object to click into, and a single visible sentence is often split across several separate drawing instructions.

Fonts add a second problem. To keep files small, the program that exported the PDF usually embeds a subset of each font that contains only the letters the document already uses. If the original text never contained a capital Q, the embedded font may have no Q to draw. Any honest editor has to check this before it writes your new words.

This is why many free tools only let you place a text box or a white rectangle on top of the page. The old words are still in the file underneath, so anyone who copies the text, searches it, or runs it through a screen reader can still find them.

Why most PDF editors charge for text editing

Reading the content stream, matching visible text to the exact operators that draw it, removing them safely and redrawing replacement text with matching metrics is real engineering work, and vendors charge for it. On Adobe's pricing page, read on 17 September 2026, “Edit text and images in PDFs” was listed on Acrobat Standard (US$14.99/mo), Acrobat Pro (US$19.99/mo) and Acrobat Studio (US$24.99/mo), each on annual billing. The same page said free Acrobat Reader covers viewing, commenting and signing, and that full editing needs a paid subscription.

Server-based online editors have a second cost: your file is uploaded, processed and stored on their machines, and they pay for that compute. The TryDocsy editor runs in the browser tab instead. The PDF is parsed with pdf.js and rewritten with pdf-lib on your own device, so there is no per-file server bill to recover through a subscription or a daily limit.

Step by step: change existing text in a PDF

1. Open Edit PDF and drop your file on the upload area, or click to browse. The uploader accepts PDFs up to 100 MB.
2. Wait for the pages to render. Text the editor can change is detected line by line from the PDF's text layer.
3. Click the words you want to change. An inline box opens with the current text. Type the correction, then press Enter to save it or Escape to cancel.
4. To change a word that appears many times, such as a company name, open Find and replace (Ctrl+F, or ⌘+F on a Mac), type both values and choose Replace or Replace all.
5. Use the other tools in the toolbar if you need them: add a text box for new wording, add an image or signature, draw shapes, or redact an area.
6. Press Verify if you want the editor to rebuild the file in memory and report whether the original wording of each edit is really gone.
7. Press Download. The file is saved with “-edited” added to its name, and a message appears if any edit couldn't be applied.

What happens to the fonts

When you download, the editor looks for the font program your document embedded for that line. If it can read the program and the program contains a glyph for every character of your new text, it redraws the line in that same font, so the correction matches the words around it. If you switch a line to bold or italic, it uses a standard font instead, because an embedded font is a single face and can't create a bolder version of itself.

If the embedded font can't be used (it is missing a letter you typed, or it is in a format the parser can't read), the editor falls back to the nearest standard PDF font: Helvetica, Times or Courier. That fallback is chosen to match the original style, but it is not the same typeface, so a heading set in a brand font may look slightly different after the edit. Standard fonts only cover Western European characters, so replacement text in Hindi, Tamil, Chinese or emoji can only be drawn when the document's own font already contains those glyphs.

Is the old text really removed?

For each edited line the editor tries to delete the original drawing instructions from the page. It only does so when it can match the line to those instructions exactly and unambiguously. When it can't, for example with an unusual font encoding, it falls back to painting a cover over the old words and drawing the new ones on top. The page looks right either way, but in the fallback case the old wording is still inside the file.

You don't have to guess which case you got. After a download that used the fallback, the editor tells you that the edit is “shown correctly, but the original wording behind it may still be found by search or copy”. Verify gives a per-file count of edits fully removed versus covered. If the old wording is sensitive, such as a salary figure or an account number, use the redact tool on that area instead, which is designed to black out and remove text.

Scanned PDFs, forms and other special cases

• Scanned pages: a scan is a picture of text, so there is nothing to click. When a file has no text layer, the editor shows “This PDF looks scanned” and offers Make scanned text editable (OCR), with a document-language menu (English, Hindi, Spanish, French, German, Italian, Portuguese, Russian, Chinese (Simplified), Japanese and Korean). OCR text is an estimate, so proofread it. To make a scan searchable without editing it, use OCR PDF instead.
• Coloured or photo backgrounds: when a line sits on a coloured or image background, the editor warns that erasing it may leave a faint seam. Zoom in on the downloaded page before sending it.
• Fillable forms: detected form fields can be filled in, and the Flatten toggle bakes the values into the page so they can no longer be changed.
• Accidental reloads: your typed edits (not the PDF itself) are kept in the tab's session storage, so reopening the same file in the same tab offers to restore them. Closing the tab clears them.

After editing: size, signature and sharing

Edits add a small amount of data, and a reused font is embedded again, so an edited file can come out slightly larger than the original. If a portal rejects it for size, Compress PDF can shrink it, but know the trade-off first: that tool redraws every page as a JPEG picture, so the text in the compressed copy can no longer be selected, searched or edited. Keep the uncompressed edited file as your working copy and send the compressed one only where size matters more than selectable text.

Sign last as well. Adding a signature and then changing text above it looks careless, and some recipients compare versions. If you must also merge the edited file with other documents, edit each file first, then combine them.

Frequently asked questions