How to make a PDF text-searchable for court

Updated

A text-searchable PDF has real text that can be searched, selected and read aloud by screen readers, not just a picture of the page. Several courts require it, and judges' staff rely on searching through filings.

The best fix: don't scan

If you wrote the document, save it directly as a PDF: in Word, File › Save As › PDF; in Google Docs, File › Download › PDF. The result is searchable, sharp, and a fraction of the size of a scan. The Tenth Circuit calls these "native PDFs" and requires them for pleadings and briefs.

For paper documents: OCR

Optical character recognition (OCR) reads the text in a scanned image and adds an invisible text layer on top of it. The page looks the same, but the text becomes searchable. California Rule of Court 2.256(b)(3) treats standard commercial OCR as enough to meet its searchability requirement.

  • Adobe Acrobat: Scan & OCR › Recognize text › In this file.
  • Many scanners have a "searchable PDF" option in their scanning software. Turn it on before scanning.
  • Docket Prime does it for free in your browser: add your PDF to the checker below and choose "Add searchable text". Only the scanned pages are processed, the original scan images are kept, and the file never leaves your device.

Pages that fool simple checks

Documents downloaded from PACER carry a header like "Case 1:23-cv-01234 Document 12 Filed…" in their text layer, even when the page itself is a scan. A simple "does it have any text?" check is fooled by that. Docket Prime also looks at how much of each page is covered by an image, so stamped scans are still flagged.

Courts that require searchable PDFs

Check your PDF now

Stays on your device

Required. Pick the court you're filing in: size limits and rules differ from court to court. Type a state, city or court name.

Filing in a state or county court, not a federal one? Find your state's courts

Drop your PDF here

Choose your court first (step 1), so we know which rules to check.

Add several files at once for exhibits. The first file is treated as the main document.

Checked and fixed inside your browser. Your files are never uploaded or stored.

Common questions

How do I know if my PDF is text-searchable?
Open it and press Ctrl+F (Cmd+F on a Mac), then search for a word you can see on the page. If nothing is found, or you can't select the text with your cursor, the page is an image. The checker on this page tests every page and lists those without text.
Which courts require text-searchable PDFs?
Among the courts we track: Alameda Superior Court, Butte Superior Court, California Superior Courts, Fresno Superior Court, Idaho state courts, Imperial Superior Court, Kern Superior Court, L.A. Superior Court, Nebraska appellate courts, NY App. Div. (2nd–4th Depts.), NY App. Div., 1st Dept., New York NYSCEF, Orange County Superior Court, Oregon state courts, Riverside Superior Court, San Bernardino Superior Court, San Diego Superior Court, San Francisco Superior Court, San Mateo Superior Court, Santa Barbara Superior Court, Sonoma Superior Court, Tehama Superior Court, Texas state courts, 10th Circuit, Ventura Superior Court. Many others recommend it.
Will OCR make my file bigger?
Only slightly when the text is added as an invisible layer, as Docket Prime does: a few kilobytes per page. Tools that re-save the scan images can make files much bigger, which is why some courts (such as S.D. Texas) advise against OCR for large scans.
Is a PDF made by "Print to PDF" searchable?
Yes, when you print from the original program (Word, a browser, an email). It is not searchable if you print a scanned image to PDF: the result is still a picture of text.