The phrase "edit a PDF" promises more than most browser tools deliver, and the gap between the promise and the reality causes most of the disappointment people feel with PDF editing. When you edit a document in a word processor, you change the words themselves: delete a sentence and the following text reflows to fill the gap, insert a paragraph and everything below it moves down. A PDF does not work that way. Its text is positioned on the page, not flowed, so there is nothing to reflow. What browser PDF tools do instead is add new content on top of the existing page: text, shapes, lines, and images placed as overlays, while the original content stays exactly where it was.
That is not a lesser version of editing. For many jobs, an overlay is exactly what is needed. But knowing the boundary beforehand prevents a painful misunderstanding: promising a client a rewritten PDF and discovering the tool cannot deliver a rewrite. Edit PDF overlays text stamps, rectangles, lines, and images onto pages, placed by clicking a live preview, processed in your browser. This post maps that model honestly, so you can tell in advance whether your job fits it.
What overlays do well
Overlays excel at additions. A missing label, a corrected figure covered and replaced, a stamp marking a draft, a logo on a first page, a signature line, a short note pointing elsewhere: all of these are additions to a page, and none of them requires the original text to move. Because the overlay sits on top, you can place it precisely, check it against the surrounding content in the preview, and download a page that reads as a finished whole. For documents whose source files are lost, overlays are often the only practical fix short of rebuilding the document, and rebuilding is rarely proportionate for a one-line change.
Overlays also preserve the original. Nothing underneath is altered, so the document's history survives in the file. For a newsletter correction, a labelled diagram, or a stamped draft, that preservation is harmless or even useful. The overlay model is additive, not destructive, and that gentleness is a feature when the change is small and the document is otherwise sound.
What overlays cannot do
The limits follow directly from the model. An overlay cannot reflow a fully typeset PDF like a word processor. If you add a paragraph to a dense page, the existing text does not move aside to make room; your paragraph sits on top of whatever was already there, obscuring it. If you need to insert a section into the middle of a report and have everything after it shift down cleanly, you need the source document, not an overlay tool. Pretending otherwise produces a page where two texts overlap, which is worse than the unedited original.
An overlay also cannot change the original words in place. Covering a wrong word with a rectangle and writing the right word on top looks correct on screen, and for many purposes it is correct enough, but the original word remains in the file beneath the cover. That matters in two situations. First, where the document's history matters, such as a contract or a financial record, a covered-over change is not a clean correction; the document should be reissued. Second, where privacy matters, a cover is not a removal: sensitive text under an overlay can still be found in the file, so overlays must never be used to hide information that must stay private.
The scanned-page limit
Scanned pages have their own, stricter limit. A scanned page is a photograph of a page, not a page of text. There are no words for an overlay tool to work alongside, only pixels. You can still place labels and marks on a scanned page visually, because the preview shows the image, but you cannot edit the scanned words as text, correct a misspelling in the scan, or reflow scanned lines. Those operations need text first, and text comes from OCR: a process that reads the image and adds an invisible, searchable text layer to the PDF.
The order matters. Run OCR first, so the scan gains a text layer, and then consider what the document needs. Even after OCR, though, the overlay limit remains: you are adding content on top of the page, not rewriting the scanned text underneath. If a scanned document needs substantial rewriting, the honest route is to retype or re-source it, not to overlay corrections onto every line. Overlays are for small, visible additions, and a page that needs rewriting line by line has outgrown them.
Worked example: a fictional price list, fixed honestly
The example is fictional: an invented business, invented prices, and an invented page. Imagine the fictional Fernwood Plant Nursery, contact via hello@example.org, whose one-page printed price list has three problems: one price is out of date, the list has no "prices include tax" note, and the nursery's new logo is missing from the top. The source file was made by a former employee and cannot be found.
The nursery works on a copy. The old price is covered with a background-matched rectangle and the new price placed on top in matching size. The tax note is added as a short line of text beneath the price table, aligned to the table's left edge. The logo image is placed in the top corner, sized to match the heading rather than dominate it. A test print shows a clean, usable price list, ready for the shop counter the same day. The nursery also notes, correctly, what it did not do: it did not rewrite the price list, it did not change the document's underlying text, and it did not hide anything sensitive under a cover. Had the price change been part of a formal record, or had the old price been commercially sensitive, the nursery would have reissued the list instead. Knowing which situation you are in is the judgement the tool cannot make for you.
Deciding whether an overlay fits your job
Ask three questions before you start. First, is the change an addition, or does it require existing text to move? If text must move, you need the source file. Second, does the document's history matter, or must anything be truly removed? If either answer is yes, reissue or redact rather than overlay. Third, is the page scanned, and if so, has it been through OCR? If it is scanned and unprocessed, overlay only labels and marks, and expect no text editing at all.
When all three answers point to overlay, proceed with confidence. Place additions by clicking the preview, align them to the page's existing structure, size them to match surrounding text, and check a test print when the document will be printed. When any answer points away from overlay, do not force the tool to fit the job. The fastest fix that misrepresents a document is not a fix; it is a future problem with your name on it. For comparing versions before and after an overlay, Text Compare handles plain-text comparisons, and OCR PDF adds the searchable layer scanned pages need before text work begins.
Honest limitations
Overlay editing cannot reflow typeset text, cannot rewrite original words in place, cannot edit scanned pages as text without OCR, and cannot delete content it covers. Those four limits define the tool, and every other claim about it should be read against them. Within those limits, overlays are fast, precise, and genuinely useful for the small fixes that make up most day-to-day PDF work. Outside them, the right tools are the source document, a reissue, or redaction, and choosing those is not a failure of the overlay tool. It is good judgement about which tool fits which job.
FAQs
Can I edit a PDF like a Word document in a browser?
No, not in the reflow sense. Browser PDF editing adds overlays, text, shapes, lines, and images, on top of existing pages. It does not move existing text aside or rewrite the original words. For small additions and corrections, overlays are ideal; for a rewrite, you need the source document.
What is the difference between editing a PDF and annotating it?
Annotation adds comments and highlights that readers can often toggle or remove, sitting alongside the content. Overlay editing adds content onto the page itself, as part of what the reader sees. Neither rewrites the original text, and neither deletes anything it covers.
Can I correct a scanned PDF with overlay editing?
Only visually, with labels and marks. A scanned page has no text layer, so its words cannot be edited as text until OCR adds one. Even after OCR, overlays add content on top rather than rewriting the scanned words beneath.
When should I not use overlay editing?
When existing text must reflow, when a correction must be a clean reissue for the record, or when content must be permanently removed rather than covered. In those cases, reissue the document or use true redaction instead. For a general guide to PDF workflows, see how to merge and split PDFs.
human review pending client designation