How to Compress PDF Files for Email

Before compressing, identify the dominant byte source. In most over-sized PDFs, a single category accounts for 70-90% of the file: embedded raster images, embedded font subsets, or incremental save history. Compressing the wrong category wastes time and quality. Open the file in any PDF inspector — Acrobat's Save As Optimized > Audit Space Usage, or `mutool info -X file.pdf` from the command line — and the breakdown is usually obvious within seconds.

Embedded images are the most common offender. PDFs store images at their native resolution regardless of the page size they appear at, so a 4000x3000 photo placed into a 600-pixel-wide column still carries the full 12-megapixel payload. The fix is downsampling: re-encoding images at a target DPI matched to viewing context. Ghostscript's `pdfwrite` device exposes this directly through `-dColorImageResolution=150` and the matching `-dGrayImageResolution` and `-dMonoImageResolution` parameters documented in the Ghostscript pdfwrite reference.

Font embedding adds bulk only when subsetting is disabled or when the document references many typefaces. A fully-embedded font family with regular, bold, italic, and bold-italic faces typically adds 800 KB to 1.5 MB. Subsetting cuts this to 50-150 KB by embedding only the glyphs actually used. Word's PDF export subsets by default, but PDF generators built on legacy printer drivers often do not — re-saving through a modern engine usually fixes this without any visible change.

Scanned PDFs are a different problem class entirely. Each page is a full-page raster image, typically 3-8 MB at 300 DPI in 24-bit color. A 20-page scanned contract weighing 60-160 MB is unfortunate but expected. For these documents, the leverage point is choosing scan resolution at capture time and converting to JBIG2 or CCITT Group 4 for monochrome text pages — both natively supported by `pdfwrite`. Once a scan has been embedded at high DPI, recovering quality is impossible; you can only re-downsample.

The four presets exposed by most online compressors — Screen, Ebook, Printer, Prepress — are not subjective quality tiers. They are named values for `-dPDFSETTINGS` in `pdfwrite`, each binding a specific set of resolution caps and image filters. Knowing the actual numbers behind each preset removes the guesswork. `/screen` downsamples color and grayscale images to 72 DPI with JPEG quality 0.5, `/ebook` to 150 DPI at quality 0.5, `/printer` to 300 DPI at quality 0.76, and `/prepress` to 300 DPI at quality 0.88 with color profile preservation. The full table is documented in the Distiller parameters reference.

For email-bound documents that will be read on screen and occasionally printed in draft mode, `/ebook` is the correct default. Its 150 DPI cap produces images that are visually indistinguishable from originals at typical viewing distances on a laptop or phone, while cutting the average illustrated report by 50-70%. `/screen` is genuinely aggressive — at 72 DPI a printed page shows visible blur, and at JPEG 0.5 fine text rasterized inside images becomes unreadable. Use `/screen` only when the recipient explicitly does not need to print.

Lossless restructuring is a separate axis. Ghostscript and qpdf both perform object stream compression, duplicate-object detection, and content-stream Flate recompression — operations that produce identical visual output but typically save 5-20% on bloated files. A PDF that has been edited, signed, and re-saved several times often carries an incremental save history (the trailing `%%EOF` markers reveal it) that linearization or `qpdf --object-streams=generate --linearize` will collapse. Try a lossless pass first; you may not need lossy compression at all.

Expected compression varies wildly by content. Image-heavy reports compress 50-80% at `/ebook`. Text-and-vector PDFs (think LaTeX-generated papers) often compress only 5-15% because their content streams are already optimal — the bytes are mostly typeset glyph references. If your text-heavy 30 MB PDF refuses to shrink, the bloat is likely fonts or embedded files, not images.

Open the Compress PDF tool and drop your file onto the upload area or click to browse. The tool accepts files up to 50 MB. Once the file uploads, you will see the original file size displayed alongside a compression level selector. For email, choose the Ebook or equivalent medium compression level, which provides the best trade-off between file size and visual quality.

Click the compress button and wait for processing to complete. The tool compresses the document server-side using Ghostscript, re-encoding images, removing redundant objects, and optimizing internal streams. When finished, the new file size is displayed so you can immediately compare it to the original. Download the compressed file and open it to verify that text remains sharp and images are acceptable for your purposes.

If the compressed file is still too large for email, try stepping down to a more aggressive compression level. If the document is primarily scanned pages, even Ebook compression may leave it above 20 MB, in which case you should consider the alternative strategies described in the next section. As a final check, try attaching the compressed file to a draft email in your mail client to confirm it falls within the size limit before sending.

When compression alone cannot bring a PDF below the email size limit, splitting the document into smaller parts is the most straightforward alternative. The Split PDF tool lets you divide a document by page range, so you can send a 40-page report as two 20-page attachments in separate emails or in the same email if each part fits within the limit. Label the parts clearly in the filename, such as "Report-Part1.pdf" and "Report-Part2.pdf," so the recipient can reassemble them easily.

For documents where the recipient only needs a visual reference rather than editable content, converting to images can reduce the file size substantially. The PDF to Image tool exports each page as a JPEG or PNG. JPEG images of PDF pages are often much smaller than the equivalent PDF page, especially for scanned documents, because you can control the image quality and resolution independently. You can then attach the images directly or zip them into a single archive.

Another approach is to remove unnecessary pages before sending. Many PDFs contain cover pages, blank pages, appendices, or sections that are not relevant to the recipient. The Organize PDF tool lets you delete, reorder, and select specific pages, producing a trimmed document that may fit within the email limit without any compression at all. This is especially effective for long reports where you only need to share a specific section.

Finally, consider whether email is the right delivery mechanism at all. For files above 20 MB, uploading to a cloud storage service like Google Drive, Dropbox, or OneDrive and sharing a download link is often faster and more reliable than email attachments. The recipient gets the full-quality file without size constraints, and you avoid the risk of the email bouncing or being caught by spam filters that flag large attachments.

The most effective way to keep PDF sizes manageable is to control the input quality before the document is created. When scanning physical documents, choose a resolution appropriate for the final use. For documents that will only be viewed on screen or sent by email, 150 DPI is sufficient and produces files roughly one-quarter the size of 300 DPI scans. Reserve 300 DPI for documents that will be printed at high quality. If your scanner offers a grayscale mode for text-only documents, use it instead of color to cut file size by another 50 to 70 percent.

In digital workflows, avoid embedding full-resolution photographs when the document does not require them. Resize and compress images in an image editor before placing them into your document. A 1200-pixel-wide image is more than sufficient for a full-page illustration in a PDF destined for email. If you are using Word, PowerPoint, or another office application, look for a "Reduce file size" or "Compress pictures" option before exporting to PDF. These features strip metadata and downsample images automatically.

Replace scanned signature pages with digital signatures whenever possible. A scanned page adds several megabytes as a raster image, while a digital signature adds virtually nothing to the file size. Similarly, avoid pasting screenshots of tables or text into PDFs; keep that content as native text and vector graphics for dramatically smaller files. By building these habits into your document workflow, you will rarely need to compress PDFs after the fact.