How to Tenthline a Scanned Legal Document

A scanned document is fundamentally different from one exported straight out of Word. A typed PDF has real, selectable text with known positions on the page. A scanned pleading, whether it came from a photocopier, a phone camera, or an old typewritten record pulled from a physical file, is really just a picture of a page. There's no text underneath for a computer to count lines from until something reads the image and figures out where the words actually are. Here's what that process looks like and how to get it right.

Why scanned documents need a different step

Numbering every tenth line accurately depends on knowing where each line of text actually sits on the page. For a typed PDF, that information already exists in the file. For a scanned PDF, it doesn't exist until the image has been read and converted into positioned text, a process called optical character recognition, or OCR. Skip this step, or do it badly, and any line numbering applied afterward is guesswork rather than an accurate count.

Step 1: Upload the scanned PDF like any other document

You don't need to run OCR yourself first or use a separate workflow. Tenthliner classifies each page during processing and automatically detects when a page is image-based rather than text-based. When it finds an image-based page, it runs that page through OCR to build an accurate, positioned text layer before applying line numbering, all within the same upload-and-process flow you'd use for a typed document.

Step 2: Check the scan quality before you upload, not after

OCR accuracy depends heavily on how clean the source scan is. Before uploading, check that:

• The page is scanned straight, not rotated or skewed
• The resolution is high enough that text isn't blurry when zoomed in
• The contrast is reasonable, dark text on a light background, not a faded photocopy of a photocopy
• Pages are right-side up and in the correct order

A crooked or low-resolution scan gives OCR less to work with, and that shows up as less reliable line detection. If you have the option to rescan a poor-quality page before filing, it's worth doing before numbering rather than after.

Step 3: Choose your preset and settings as usual

Once the document is uploaded, pick the court preset or custom settings (font, page numbering position) the same way you would for a typed document. The OCR step happens automatically during processing and doesn't require any separate configuration from you.

Step 4: Allow a bit more processing time

Because a scanned document goes through an additional OCR step before line numbering can be applied, processing can take longer than it would for a typed PDF, particularly for longer documents. This isn't a sign of a problem. It's the extra work being done to build an accurate text layer before counting lines against it.

Step 5: Download and spot-check the result

Once processing completes, open the finished PDF and check a few pages against the original scan, particularly any pages with unusual layouts, tables, or handwriting mixed with typed text. OCR is very reliable on clean typed text and less reliable on handwriting, so a document that mixes both is worth a closer look before you file it.

When this matters most

This comes up most often with older case files, historical records being digitized for the first time, or documents received from a client or opposing party as a physical copy rather than a native digital file. In all of these cases, the underlying content still needs to meet the same formatting requirements as any other filing: page numbers in the right corner, every tenth line marked where the applicable practice direction requires it. The scanned starting point doesn't change what's required. It just means an extra, automatic step happens before the numbering can be accurately applied.

For the numbering step itself on documents that already have real text, see How to Number Every Tenth Line in a PDF Without Retyping. For the underlying formatting rules by court, see the Supreme Court, Court of Appeal, and High Court breakdowns.