Checked against the current Windows build. The public download is still pending.
First, find out whether the page holds text at all
Open the file, go to the Edit tab and double-click the words you want to change. If a small editing box opens over them, the page stores real text and you can retype it. If nothing happens, you are looking at a picture of a page — a scan, or a photo somebody dropped into a PDF — and no editor will let you retype that.
For a scan, run OCR first. That makes the words findable and copyable. It does not make them editable: the recognised words go underneath the image, invisibly, and the image itself never changes. If the blanks you are trying to fill are real form fields, the form-filling workflow is a different and far better-behaved tool.
One line, or the whole paragraph
Edit text replaces a single run of glyphs and leaves everything else exactly where it was. Shorter text leaves a gap; longer text runs past the margin. That is not a bug, it is what a PDF page is, and every editor that appears otherwise has a reflow guess sitting on top of the same thing.
Ours is Edit on page: double-click and the whole block opens, re-wrapped to its own width. The block is a guess from geometry — same font, same size, even baselines, a shared left edge — so a table with tidy rows can read as one paragraph, and a paragraph that changes font halfway reads as two. Untick Whole paragraph when you want the single line instead.
Read the three warnings before you save
Add a sentence and the block may need a line it does not have. The re-wrap happens inside the block and moves nothing else on the page, so it grows downwards over whatever sat below — and it says so, in words, every time. A word too long to break reaches past the right edge, and says so. A character the chosen face has no glyph for is dropped silently by the renderer, so the finished page is read back and compared with what you asked for, and anything that did not arrive is named.
None of the three stops the edit. They tell you where to look. Zoom in on the last line of the block, the line under it, and any label sitting to its right. Then Save, or Save a copy if somebody else still needs the version you started with — nothing has touched the file on disk until you pick one.
Fonts decide how much survives
If the original run used one of the standard fourteen faces — Helvetica, Times, Courier and their bold and italic versions — your text is re-typed in place and nothing changes but the words. If the font is embedded it is usually a subset holding only the glyphs the author actually typed, so a letter they never used would draw as nothing. That run is set again in a Windows system font of the same class, and the result names the substitution.
Tabs and newlines are refused rather than quietly turned into spaces: a tab usually means somebody wanted alignment, and a silent space is wrong in a way nobody goes back to check. And if the real job is rebuilding a layout rather than fixing a sentence, go back to the source — Word export recovers the words and how they were set, not the design.
Common questions
Can I edit a scanned PDF the way I edit a Word file?+
No. OCR puts searchable words under the scan, but the scan is still a picture. To change what the page says you need the file it was printed from.
Does an edit touch my file straight away?+
No. Edits sit in the session until you press Save. Save a copy writes a new file and leaves the original alone.
The tools this guide uses.
Open the complete 44-tool catalogue, review the current limitations, or check Windows release availability.