Edit a PDF header or footer on every page, in 1 pass

Fix the header on every page

Edit PDFReal text editing, objects, annotations, forms. Free, runs on your device.Open

1. Drop the PDF on the editor above, or open /edit.
2. Press Ctrl+H. The find bar opens with the replace row showing. On a Mac, press Cmd+F and click Replace… in the bar.
3. In Find in the document, type the part of the header that is wrong, exactly as it appears: “Draft 0.9”. The counter shows how many pages carry it, for example “1 of 48”.
4. In Replace with, type the new text: “Final 1.0”.
5. Choose Replace all. Every match changes in place, as real text in the header’s own position.
6. Check (next section), then save with Ctrl+S. The file downloads with “-edited” in its name.

On our test report the engine replaced all 48 headers in well under a second on a laptop (0.3 seconds when we measured it again). After saving, a search for “Final 1.0” found 48 matches and a search for “Draft 0.9” found none. Replace all is 1 step in the history, so 1 Ctrl+Z puts all 48 back.

Type only the part that changes. Our header read “Annual report 2026, Draft 0.9”; searching for “Draft 0.9” alone leaves the rest of the line exactly as it was, so there is less to go wrong.

Check every page in 10 seconds

You do not need to scroll through 48 pages:

1. Search for the new text. The counter should match the number of pages that had the header: “1 of 48” on our report.
2. Search for the old text. The counter should say “No matches”.
3. For a visual check, look down the thumbnail rail: headers and footers sit at the edges of each thumbnail, where a missed or overflowing line shows.

If the new text is longer than the old, it runs on to the right from where the line starts; it does not wrap. A header that was centred grows to the right, not from the middle. Step through a few hits with Enter and look.

When the line differs on each page

Page numbers and per-page dates are not 1 repeated string, so 1 search cannot match them all. What still works:

What you need	What to search for	Result on our 48-page report
“of 48” to “of 50” after adding 2 pages	of 48	48 matches, 1 per page
Only page 3’s footer	Page 3, with ab (whole words only) on	1 match
The same, without whole words	Page 3	11 matches: pages 3 and 30 to 39
A date that is the same on every page	the date, as printed	1 per page
A different date on every page	the shared part, such as the year	step through with Enter and replace each

Find matches exactly what you type; there are no wildcards or patterns. For a handful of stragglers, double-click the line on that page and retype it, which is often faster than building the perfect search.

Page numbers do not renumber themselves. To the file, “Page 3 of 48” is plain text on page 3, not a field. Delete page 2 in the thumbnail rail and every later page keeps its old number, so plan page changes first and fix the numbering last.

Headers that are pictures or grouped

Not every header is text you can retype:

• A logo strip or a scanned letterhead is a picture on each page. Find cannot see it. Select it on each page and press Delete, or add the new logo with the Picture tool and copy it across pages with Ctrl+C and Ctrl+V, which keeps the same position. It is page-by-page work; on a long document, that is the honest cost.
• A header grouped into 1 object by the authoring software moves as a block, and its text cannot be retyped. Find still finds the words, and Replace all skips them and says how many “could not be changed here, because the text sits inside a group, is written letter by letter, or is an OCR layer”. Cover those headers with a white rectangle and add new text, or ask for the source file.
• A header turned into outlines has no characters left; nothing finds it. Cover it and add new text, or ask for the source file.
• A scanned document has no text anywhere until OCR adds an invisible layer, and even then the visible header is the scan’s ink. Edit a scanned PDF covers the cover-and-retype method.

Adding a header or footer where there is none

That is a stamping job, not an edit, and Add page numbers does it for the whole file:

1. Drop the PDF on Add page numbers.
2. Under Where, pick a place for the numbers, or No numbers if you only want a header or footer.
3. Pick a Format: “1”, “Page 1”, “1 of 12” or “Page 1 of 12”.
4. Type a Header (top centre) or a Footer (bottom centre). The placeholders {page}, {pages}, {date} and {file} are filled in on each page; {date} comes out as year, month and day, such as 2026-10-11.
5. Set Cover page to “skipped” to leave page 1 bare, then run it.

The stamp is real text in Helvetica, drawn on top of each page. It adds; it never removes. When we stamped “Page 1 of 48” numbers on our report, which already had its own, page 3 showed “Page 3 of 48” twice. Fix or remove an existing header in the editor first, then stamp. The stamp tool uses the standard PDF font, so letters such as ł or ř are refused with a message; add those in the editor, which embeds a font for them. More page jobs live in the page organising guide.

Where this stops

Find and replace changes text, not layout. It cannot move a header from the left to the centre, change its size for all pages at once, or rebuild a footer as a Word-style section footer. A changed header on a digitally signed file is appended after the signed bytes, which keeps the signature valid for the signed version but leaves the old header inside the file; the editor says so when it saves. And if you have the Word file the PDF came from, changing the header there and exporting again is the cleaner fix. For everything else about changing text, see the find and replace guide and the editing guide.