BASALT · JOURNAL
The best PDF software for lawyers working on a Mac
The best PDF software for lawyers is not the one with the longest feature list. It is the one that fails safely, because in a law practice the failure mode of a PDF tool is a disclosure, a sanction, or a filing that gets kicked back.
This is a way to evaluate any candidate against the work a practice actually does, rather than against a comparison chart.
What a practice actually does to PDFs
Strip away the marketing and the operations are fairly narrow, and they repeat constantly.
Documents arrive as scans and have to be made searchable. Exhibits get assembled, split, reordered, and stamped with sequential numbers. Productions get redacted and logged. Drafts get compared. Files get compressed to fit an e-filing size limit. Nearly everything eventually gets flattened and sent.
Notice what is not on that list: rewriting body text. Lawyers edit in a word processor and produce PDFs from it, so the PDF work is assembly, sanitization, and proof. Software chosen for text-editing prowess is chosen against the wrong criterion.
Criteria for the best PDF software for lawyers
Redaction that destroys rather than covers. A PDF page is a content stream of drawing operators, and text is drawn by operators that take strings of character codes. A black rectangle is another drawing operator painted on top; the codes stay in the file and remain extractable with copy and paste. Real redaction removes the operands and re-emits the surviving text with corrected positioning. PDF redaction for law firms covers it in depth.
Proof, not assertion. Most software reports success when its code path finished without an error, which is a program grading its own homework. The stronger design re-opens the written file with an independent parser and refuses to save if the removed content is still findable. When a production is challenged months later, the difference between "we used the redaction tool" and "here is the verification record" is the difference between an argument and an answer.
Hidden data handling. Removing text from a page does not remove the author name in the metadata, the original filename in the XMP packet, the quoted passage in an annotation's contents, a form field value, an embedded attachment, or a layer that is switched off rather than absent. It also leaves prior versions of objects behind when the save is incremental. A tool must sweep all of it in one operation, not as a separate command someone has to remember.
Bates numbering that survives. Sequential numbering has to be applied consistently across a set, positioned so it does not collide with existing content, and it has to remain intact through later operations. Bates numbering on Mac covers prefixes, ranges, and placement.
OCR for scanned discovery. Scans are pixels. Without a text layer they are neither searchable nor reviewable at volume, and redacting them means destroying image data rather than editing text. OCR on Mac explains the limits.
Confidentiality as architecture. If a tool uploads documents to a server, that transfer is a disclosure of client material to a third party, and it happens before any processing. Whether it is permitted depends on your engagement terms, any protective order, and your jurisdiction's confidentiality rules. A local tool avoids the analysis.
Originals preserved. Every operation should write a new file. In-place editing destroys your ability to re-review a source document, and it destroys the chain you would want to show.
Size control for e-filing. Courts impose per-document size limits. Compression must be controllable and must not silently destroy the OCR layer or the numbering.
A log you can hand over. Redactions carry reasons. Producing a privilege log from the redactions themselves rather than reconstructing it by hand removes a category of transcription error.
The options, honestly
Apple Preview is free and on every Mac. It views, reorders, merges, splits, signs, and annotates competently, and a great deal of routine work never needs more. It is not appropriate for redaction: its markup tools draw rectangles, and it saves PDFs as incremental updates that leave prior object versions in the bytes.
Adobe Acrobat Pro is the standard in litigation support and the tool most vendors and opposing counsel assume. Its redaction performs real content removal and it covers essentially everything a practice needs. It is a subscription, it is a large install, and sanitizing hidden data is a separate command from redaction, which makes the order of operations a human dependency.
Browser services such as Smallpdf and iLovePDF are fast and convenient for merging and compressing. Sending client documents through them is a confidentiality question you have to answer before you use them, not after.
Native Mac editors in the PDF Expert and PDFpen category are capable for assembly, annotation, and forms. Their legal-specific features vary considerably, and their redaction should be tested rather than assumed.
Litigation support platforms are a different category: hosted review, production, and coding at volume. If you run large productions that is the right tool, and a desktop PDF app is not competing with it. Most practices need both.
Basalt is a native macOS app built around the sanitization end of this list. It is $29 once for up to three Macs, free for 24 hours.
What software will not solve
No tool decides what is privileged. No tool knows that a redacted name appears again, unredacted, deep in an exhibit nobody re-read.
Software can make removal reliable and provable. The review remains a human obligation, and it is worth running a fixed checklist before filing every time rather than trusting recollection.
Frequently asked questions
What is the best PDF software for lawyers on a Mac?
The best option is whichever tool redacts by removing content rather than covering it, sweeps hidden data in the same operation, applies Bates numbering reliably, OCRs scans, and keeps documents on your machine. Acrobat Pro is the broadest and is subscription based; native Mac apps such as Basalt cover the sanitization and assembly work under a perpetual license. Test redaction on a real document before committing.
Do lawyers need Adobe Acrobat?
Not necessarily, though many firms standardize on it because vendors and opposing counsel assume it. The operations a practice depends on, meaning OCR, page assembly, Bates numbering, compression, signing, and redaction, are all available in native Mac applications. Acrobat's clearest advantage is breadth and its position as the default in litigation support, not a capability that cannot be replaced.
Is Preview safe for redacting client documents?
No. Preview's markup tools paint a rectangle onto the page without removing the text-showing operators underneath, so the characters remain selectable and extractable by anyone who receives the file. Preview also saves PDFs as incremental updates, leaving earlier versions of objects in the file. It is a good viewer and an unsafe redaction tool.
How do I prove a redaction was done correctly?
Re-open the produced file with a parser that had nothing to do with writing it, extract all text, and search for the terms you removed, then check metadata, annotations, form fields, and attachments separately. Keeping that output alongside the production gives you contemporaneous evidence rather than a recollection. Some tools generate a signed certificate recording the file hash.
Can I use online PDF tools for client files?
Only if you have concluded that transferring client material to a third party is permitted by your engagement terms, any protective order, and your jurisdiction's confidentiality rules. The upload happens before any processing, so the transfer is the event to analyze, not the service's output. Many practices avoid the analysis by using local software.
What is Bates numbering and does my PDF tool need it?
Bates numbering applies a unique sequential identifier to every page of a production so any page can be cited unambiguously. If you produce documents in litigation you need it, applied consistently across a set with a configurable prefix, start number, and position. It must also survive later operations such as compression and flattening without being displaced.
How do I make scanned discovery searchable?
Run OCR, which analyzes the page image and adds an invisible text layer aligned to the visible glyphs, leaving the scan itself unchanged. Accuracy depends on scan quality, so poor originals produce imperfect text and should be spot-checked. Redacting an OCR'd scan requires destroying the image pixels and editing the text layer, because either alone leaves the information recoverable.
Does redacting a PDF remove the metadata too?
Not automatically, and assuming otherwise is a common cause of disclosures. Author names, original filenames, and revision history live in the document information dictionary and the XMP packet, and text also persists in annotation contents, form field values, embedded attachments, and switched-off layers. A tool should strip all of these in the same operation.
Doing it in Basalt
Basalt covers the practice-facing work in one window: redact, OCR, Bates numbering, merge, split, organize, compress, compare, sign, inspect, and privilege log, with documents open as tabs. Redaction excises text from the content stream glyph by glyph, destroys and re-encodes image pixels, strips metadata, XMP, attachments, and hidden layers, and writes one clean generation, and a verifier proves the content is gone with an independent parser before anything reaches disk. Every job can emit a signed certificate of redaction that anyone can check with shasum and openssl. It is not a page-layout editor, it does not do certificate-based digital signatures, it is Mac only, and it will not open a password-protected file without the password. $29 once for up to three Macs, free for 24 hours, at basaltformac.com.
Redaction that proves itself
Basalt destroys the content you mark, then re-opens the file it wrote and proves the content is gone before it saves anything. Your documents never leave your Mac.