BASALT · JOURNAL
Why scanned PDFs are so much larger than text PDFs
A page of text in a PDF is a few hundred bytes: a font reference, a position, and a string. The same page scanned is a photograph, and a photograph of a letter sized page at 300 DPI in colour is several megabytes before compression.
That single difference explains almost every size surprise in document work.
The arithmetic
A letter page at 300 DPI is 2,550 by 3,300 pixels. In 24 bit colour that is about 25 MB uncompressed. JPEG at a sensible quality brings it to somewhere between 200 KB and 1 MB depending on the content, with text and line art compressing worse than photographs because sharp edges are exactly what JPEG handles badly.
Multiply by a thousand pages and a production is comfortably several hundred megabytes. Nothing has gone wrong.
What drives the number
Resolution is the biggest factor and it is quadratic. Going from 300 DPI to 600 DPI does not double the size, it roughly quadruples it, because you are doubling both dimensions.
Colour depth is the next. Colour is three times greyscale, and greyscale is many times bilevel. A document scanned in colour when it is entirely black text on white paper is paying three times over for nothing.
Compression settings matter, and this is where care is needed, because aggressive JPEG compression on scanned text produces artefacts around characters that can make a document harder to read and, in the worst cases, ambiguous.
Reducing it safely
Convert colour scans of monochrome documents to greyscale. This is usually the single largest saving available and it discards no information that was in the original document.
Reduce resolution to what is legible, not to what is minimal. 300 DPI is the normal floor for documents that may be read closely or OCRed. Below that, OCR accuracy falls off quickly.
Leave the original alone. Compress a copy. Compression is lossy and the original is evidence.
What not to do to a production
Do not re-compress a document you received as a production and intend to produce onward. Every JPEG generation loses a little more, and a document that has been through several rounds can develop artefacts that were not in the original. If size is a problem, split rather than squeeze.
Do not assume a smaller file is a cleaner one. Compression reduces the image, not the metadata, attachments, or earlier revisions the file may carry. Those are a separate question and they are the ones that leak.
Frequently asked questions
Why is my scanned PDF so large?
Because each page is a photograph rather than text. A letter page at 300 DPI in colour is about 25 MB uncompressed and a few hundred kilobytes to a megabyte after JPEG compression, so a thousand page scanned production reaching several hundred megabytes is normal rather than a fault.
Does scanning in colour make a PDF bigger?
Yes, roughly three times larger than greyscale for the same resolution, because each pixel carries three channels instead of one. A document that is black text on white paper gains nothing from colour, so converting it to greyscale is usually the largest safe saving available.
Does doubling scan resolution double file size?
No, it roughly quadruples it. Resolution applies to both dimensions, so 600 DPI has four times the pixels of 300 DPI. That is why resolution is the most effective setting to change and the easiest to overshoot.
Is it safe to compress a scanned production?
Compress a copy, never the original, and avoid re-compressing a document that has already been through a JPEG generation. Each pass loses detail around text edges. If the file needs to be smaller for delivery, splitting it is lossless where re-compression is not.
Doing it in Basalt
Basalt is a native macOS PDF toolkit with eighteen tools in one window. It opens large documents without loading them into memory, renders pages on demand, and copies files into its engine in fixed-size chunks, so peak memory follows the chunk size rather than the file size. Redaction destroys content rather than covering it, and an independent verifier re-opens every written file to prove the material is gone before the file is saved. A one time $29 licence covers up to three Macs, it is free for the first 24 hours, and the engine holds no network entitlement at all, which macOS enforces at the code-signature level. Download it 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.