Fix a PDF that's 'not text-searchable' and got rejected

Why courts require text-searchable PDFs

Courts moved to electronic filing so that documents can be searched, indexed, and read by assistive technology like screen readers. A scanned image can't be searched — it's just a photograph of a page — so most e-filing rules now require that the text in your PDF actually be text the system can read. When it isn't, the filing is rejected or returned before it ever reaches a judge.

In California trial courts, Rule of Court 2.256(b)(3) requires that an electronically filed document "must be text searchable when technologically feasible without impairment of the document's image." The rule's advisory committee comment spells out that "technologically feasible" simply means running standard, commercially available OCR software — nothing more demanding than that. California's appellate courts go further: Rule of Court 8.74 requires briefs and similar documents to be filed as text-searchable PDFs, with a narrow exception for material that can't practicably be converted, such as documents that are entirely handwritten, photographs, or graphics that aren't primarily text.

New York's e-filing system, NYSCEF, states in its filing requirements that "all PDFs shall be text searchable," and notes that scanned documents should be run through OCR software before filing. Failure to meet the specifications can cause the e-filed document to be returned.

In federal court, the picture is set by each district's CM/ECF local rules and administrative procedures rather than one national rule, but the near-universal expectation is the same: PDFs must be text-searchable, and scanned exhibits should be OCR'd before they're uploaded. Because the exact wording and file-size limits vary from district to district, check your court's own CM/ECF procedures for the specifics that apply to you.

How to tell if your PDF is already text-searchable

You don't have to guess. There are two quick self-tests you can run in any PDF viewer in under a minute.

First, open the PDF and try to select a line of text with your cursor, the way you'd highlight a sentence in an email. If the text highlights word by word, it's real text and probably searchable. If your cursor just draws a box over the whole page, or nothing highlights at all, the page is an image.

Second, use the viewer's Find command (Ctrl+F on Windows, Cmd+F on a Mac) and search for a word you can plainly see on the page. If the viewer jumps to it, you have a text layer. If it reports no results for a word that's clearly printed right there, the PDF isn't text-searchable.

A document can be mixed — some pages typed and searchable, others scanned and not. That's common when you combine a brief you wrote with exhibits you scanned. If any page fails these tests, the whole file can be flagged. docu checks every page for you and tells you exactly which ones need OCR, for free.

How OCR works, in plain language

When you scan a page, your scanner takes a photograph of it. To a computer, that photo is just a grid of colored dots — it has no idea that a particular cluster of dots is the letter "e." That's why you can't search or select it.

OCR is software that looks at that image and recognizes the shapes as letters and words, rebuilding the actual text. It then stores that text as an invisible layer sitting directly behind the picture of your page. You still see your original scan exactly as it was, but now the court's system — and search, and screen readers — can read the text underneath.

Because OCR is reading shapes, clean, straight, high-contrast scans come out most accurate. Crisp printed type is recognized very reliably. Faint copies, skewed pages, or heavy handwriting are harder, and no OCR tool is perfect on those. The good news for court filing is that the requirement is about making the document searchable with standard OCR, not about achieving a flawless transcription.

Fix it with docu, step by step

docu adds the OCR text layer in your browser. The document is opened and processed on your own device — it is never uploaded to a server — so a sealed filing, medical records, or anything sensitive stays with you the entire time. Here's the whole process:

1. 1Open the fix tool and drag in the PDF that was rejected. It loads directly in your browser; nothing is sent anywhere.
2. 2docu checks the file for free and shows you which pages are image-only and need OCR, along with anything else that might trip the portal, like an oversized file or a non-Letter page size.
3. 3Start the fix. docu reads each scanned page and builds a searchable text layer behind it, leaving the visible image of your pages untouched.
4. 4Run the two self-tests on the result if you want to confirm — select some text and use Find. It should now behave like a normal typed document.
5. 5Download your filing-ready PDF and a compliance certificate recording what was done, then upload it to your court's e-filing portal in place of the rejected file.

Doing it yourself without docu

OCR isn't unique to docu, and it's worth knowing your options. If you own Adobe Acrobat Pro, it has built-in OCR: open the PDF, go to the Scan & OCR tool, and choose to recognize text in the file. This works well but Acrobat Pro is a paid subscription.

Many desktop scanners can apply OCR at scan time — look for a "searchable PDF" or "OCR" option in your scanner software before you scan, and it will bake the text layer in as it goes. This only helps if you still have the paper and haven't already scanned it as a plain image.

Free and open-source tools exist too, such as OCRmyPDF, though they're command-line programs aimed at more technical users. Whatever route you take, the goal is identical: end up with a PDF whose text you can select and search, then run the self-tests above before you refile.

How to avoid this next time

The simplest habit: whenever you can, create your PDF by exporting or "printing to PDF" straight from Word, Google Docs, or WordPerfect instead of printing the document and scanning it back in. A PDF made directly from a word processor is text-searchable from the start and will never trip this rejection.

When you genuinely have to scan — signed documents, exhibits, older records — turn on OCR at the moment you scan, or run the scanned file through OCR before you combine it with anything else. And when you assemble a filing from several sources, remember that one scanned exhibit can make the whole combined PDF fail the check, so test the final file, not just the parts.