How to make a scanned PDF searchable (and when it works)

Updated 14 November 2026 · By the ShushPDF team · 4 min read

Quick answer

Open ShushPDF's OCR PDF, tap Choose PDF and pick your scan. Tap Make searchable. Each page is read on your device and its words are added as invisible text, so you can search, select and copy them. Then tap Download. It reads English, Sinhala or Tamil (pick one next to Text in), and works best on clean, straight scans of printed text.

A scanned PDF is a stack of pictures of pages. You can read it, but your computer cannot: searching finds nothing, and you cannot copy a sentence. OCR (optical character recognition) fixes that by reading the words in each picture. This guide shows how to do it for free in your browser, and what to expect from real scans, good and bad.

How to make a scanned PDF searchable with ShushPDF

  1. Open OCR PDF on your phone or computer.
  2. Tap Choose PDF and pick your scan. We check every page first. A scan shows "No text found in this page" (or these pages).
  3. Tap Make searchable. The first time, this downloads the text reader (about 6 MB) from our own website. Your browser keeps it for next time.
  4. Wait while each page is read. You can tap Cancel at any time.
  5. On "Your PDF is searchable", check the number of words found, then tap Download. The new file name ends in "-searchable". Copy text and Download .txt give you the words on their own.

The pages look exactly the same afterwards. The words are added as invisible text, at the place where each word is printed. If some pages already have text, you can choose Skip pages with text to leave them as they are.

How to check it worked

Open the new file in any PDF reader and search for a word you can see on the page, such as a name or a place. Pick an unusual word: if OCR read it right, the reader jumps to it. Try two or three, because one misread letter is enough for a search to miss a word.

What we found with real scans

On 13 November 2026 we made two real public scans from the US National Archives searchable, then searched them with a separate program (pdftotext).

ScanWords foundTimeSearch
Typed 1982 letter, 2 pages4864.4 to 4.9 sPlace names found
Grey 1964 routing sheet, 1 page921.8 sSome words found, some misread

The clean, typed letter worked well. The routing sheet is a grey photocopy with stamps, a form grid and handwriting. On it, "Kennedy" could not be found, and "March" came out as "Mareh". Stamps and handwriting turned into nonsense words. That is normal for OCR: it reads printed letters on a clean background best.

The times are from a computer. A phone reads one page at a time too, but will usually be slower, so a long document may take a while.

Tips for a better result

Honest limits

FAQ

Which languages does it read?

English, Sinhala and Tamil, one at a time. Pick the language next to Text in before you tap Make searchable. With English, letters with accents can come out wrong. Other alphabets are not read.

How accurate is it?

It depends on the scan. Clean, typed pages work well: on our 1982 letter, every place name we searched for was found. On a grey photocopy with stamps, some words were misread.

Is my scan uploaded?

No. Every page is read inside your browser, on your device. The only download is the text reader, from our own website, the first time.

Will the PDF get bigger?

A little. Our 2-page letter went from 115 KB to 125 KB, and the 1-page sheet from 74 KB to 76 KB.

OCR PDFMake a scanned PDF searchable, so you can find and copy its words. Free, and your file never leaves your device.
Open OCR PDF

How this guide was made: drafted with AI help, tested with scripted browser runs with ShushPDF on 13 November 2026, and reviewed by the ShushPDF team before publishing.

More guides