Why can't I search my PDF? Four causes and how to fix each

Short answer: Search only finds text the PDF stores as text. If the page is a scan (a picture), the letters were converted to shapes, or the font's character map is wrong, there is nothing matching to find. Check the file with the text-layer check: scans need OCR, shapes need OCR too, a broken map needs OCR or the original file.

Last updated:

The short answer

Search (Ctrl+F, or Cmd+F on a Mac) looks for the word in the text stored in the file, not in what you see. Four things break that link:

What you seeWhat is in the fileSearch finds it?What to do
A scanned pageA picture of the page, no textNoAdd a text layer with OCR
Normal-looking textLetters drawn as shapes (outlines)NoOCR, or ask for the original file
Normal-looking textText with a wrong character mapNo (search sees other characters)OCR, or ask for the original file
Normal textNormal textShould — check the viewerTry another viewer; check search options

Check which case you have

Open the file in the text-layer check. It reads every page in your browser (no upload) and reports per page:

  • No text found (page has images) — usually a scan; a page with an image whose letters are drawn as shapes gives the same result, so look at the page too. Search cannot work until the page has a text layer.
  • No text found (drawings) — no extractable text: the letters may be converted to outlines (they look exactly like text but are not), the page may hold only drawings, or its text may have an unusable character map. Look at the page and use the steps below to tell these apart.
  • Text found — the file stores text. Now copy a word from the page and paste it into Notepad: if it pastes as the same word, the text is fine and the problem is the search itself (next section). If it pastes as other characters, the character map is broken — see why copied PDF text turns into garbage.

Try it with our test files

We made the same delivery note four ways. They look identical on screen. Download them and search each one for 0417:

FileWhat it isText-layer check saysSearching for “0417”
text-layer-ok.pdfOrdinary textText found, 190 charactersFound
scan-only.pdfPage image, no textNo text found (page has images)Not found
text-as-outlines.pdfEvery letter drawn as a shapeNo text found (drawings)Not found
broken-text-mapping.pdfText with a wrong character mapText found, 215 charactersNot found

The last row is the confusing one: there is text, but searching for “0417” fails because the file says those glyphs are other characters (the order number comes out as 1646:1485). We ran these checks with our own tools in Chromium on 12 November 2026; you can repeat them with Search multiple PDFs, which takes all four files at once.

Fix: scans and outlined text

Both need OCR (optical character recognition), which reads the letters from the picture and adds an invisible text layer:

  1. Open Make PDF searchable and choose the file.
  2. Tick the document’s languages (English, Chinese, Japanese).
  3. Check the proposed plan. Pages with 50 or more extractable characters are set to Skip; all other pages, including pages with only a few characters, are set to Add text layer, which keeps their existing text. For a page with little text, choose Skip if its text is correct and complete, Add text layer if it is only a page number or a heading, or Redo page if the old text is wrong or partial (the page is replaced by its image and the old text layer is removed).
  4. Save the new file and search it again.

Pages that only get a text layer keep their content and look the same. A page set to Redo page becomes an image (JPEG) with a new text layer, so it can lose some sharpness and vector detail; check the saved file. OCR can misread letters, so search for a word you know is there to check. For outlined text, OCR works the same way, but asking the sender for the original file (or an export without “convert text to outlines”) gives exact text.

Fix: a broken character map

The font in the file maps its glyphs to the wrong characters, or does not say which characters they are. It happens with some fonts and some programs that create PDFs. Whether it can be repaired inside the PDF depends on what information the file still holds, and general-purpose tools rarely manage it reliably. The two ways that work for most people:

  • Ask for the original document or a new export.
  • Run OCR with Redo page in Make PDF searchable: the page becomes an image and gets a new, recognised text layer instead of the broken one.

If the text is fine: the viewer

  • Search options: “Whole words only” or “Match case” may be on; the word may be split across lines or hyphenated.
  • Search scope: the normal Find (Ctrl+F) searches the pages of the open PDF; check whether your viewer limits it to a selection or page range. To search other files as well, use the viewer’s advanced search or Search multiple PDFs.
  • Rendering problems: try another viewer (a browser, Acrobat Reader, Preview). If one finds the word and another does not, viewers handle this file differently (for example a missing or faulty character map). Use the viewer that works for now; if the file must be searchable everywhere, ask for a fresh export or rebuild its text layer.

Need the text out of a scan rather than a searchable file? See how to copy text from a scanned PDF.

FAQ

The word is clearly on the page. How can search not find it?

What you see is drawn on the page; what search reads is the text stored behind it. A scan has a picture and no stored text; letters converted to shapes are drawings; and a broken character map stores other characters than the ones drawn. In all three cases the page looks right and search has nothing to match.

Search finds some words but not others. Why?

Common reasons: the word is split over two lines or hyphenated, it is inside a picture (a logo, a stamp, a pasted screenshot), or the old OCR layer recognised it wrongly. Try a shorter part of the word. If the page has an old, poor OCR layer, redo it with Make PDF searchable.

Will converting to Word fix it?

Only if the converter runs OCR. Converting a scan without OCR gives a Word file with a picture in it, which is no more searchable than the PDF.

Sources

  1. Adobe: Searching PDFs (Acrobat help) · Checked on 2026-11-12

Desktop app for Windows

The Windows installer has not been published yet. There is no download link on this site until it is.