Fonts are where honesty matters. Text set in a font that was never embedded cannot be re-encoded without the font file, so those pages are rendered to a 200 DPI image — you can decline and keep the objection instead — and the report points to veraPDF for a certified verdict, because no browser tool can promise a formal pass. For the other two rules: OCR adds a searchable text layer to scanned exhibits in eleven languages, and Split under a size breaks a bundle into volumes that each stay below the court’s cap, measured by really writing them. Stamp Bates numbers before splitting so the sequence runs across volumes.
How it works
- Drop the filing and open PDF/A archive — the preflight appears immediately
- Choose whether pages with non-embedded fonts should be rasterized, then convert
- OCR any scanned exhibits first, and split under the court’s size cap last
Questions people ask
Will the output pass the court’s validator?
Bindery fixes the rules that cause most rejections and re-checks them, and the report lists every rule it covered so a formal validator brings no surprises. It does not claim a guaranteed pass — no browser tool honestly can — and points you to veraPDF when a certified verdict is required.
In what order should I run the steps?
Clean up and OCR scanned pages first, so the text layer exists; stamp Bates numbers next, so they run across the whole document; convert to PDF/A; then split under the size cap so every volume is conformant. A Pipeline can hold the sequence so it is repeated the same way each time.
Why does the converter rasterize some pages?
PDF/A requires every font to be embedded. Text drawn in a font the file does not contain cannot be re-encoded without that font, so the only conforming option is to draw the page as a 200 DPI image. Bindery tells you which pages are affected and lets you decline the rasterization.