Why Is My PDF So Big?

PDFs get large from embedded images, fonts, scanned pages, and unused objects. Here are the 7 most common causes — and how to shrink your file. For most documents, running the file throughCompress PDF (local, private, no upload) is the fastest first move; the rest of this guide explains what to do when that is not enough.

The 7 common causes

1. High-resolution embedded images

The single biggest cause of large PDFs. A photo placed at 300 DPI but displayed at 2 inches wide is stored at full resolution even though the reader never sees that detail. Marketing decks and product brochures are notorious for this.

How to detect: open the PDF and zoom to 400%. If images stay crisp far beyond their display size, they are over-sampled.

How to fix: re-export from the source application with downscaled images (150 DPI for screen, 200 for print). If you do not have the source, compress the structure first — it will not shrink already-compressed images, but it removes other bloat.

2. Embedded full font sets

A PDF that embeds an entire font (every glyph, including ones the document never uses) can carry hundreds of kilobytes of unused font data. This is common with older exporters or documents built from several merged PDFs that each embed the same font.

How to detect: a short, text-only PDF that is unexpectedly large (say, 500 KB for two pages of text) is a tell.

How to fix: re-export with font subsetting from the source app. Most modern exporters subset by default; older ones do not.

3. Scanned pages

A scanned PDF is essentially one large image per page. At 300 DPI, a letter-size page is roughly 8 megapixels — multiply by the page count and the file gets big fast. Scanned contracts and books are the usual suspects.

How to detect: try to select text. If you cannot, the page is an image. The file size will also scale roughly linearly with page count.

How to fix: re-scan at a sensible DPI (150–200 for text) before anything else — this is the biggest win for scanned documents. Thencompress the result.

4. Complex vector graphics

Detailed vector art (CAD drawings, maps, intricate illustrations) can be large even though it is not raster. Thousands of path segments and gradients add up. Unlike images, vectors do not have a "DPI" to turn down.

How to detect: the PDF is large but contains no embedded photos and no scanned pages.

How to fix: simplify the artwork in the source application, or flatten complex vectors to raster at a chosen DPI if the recipient only needs to view the file.

5. Embedded attachments and portfolios

PDFs can contain other files embedded as attachments — source documents, spreadsheets, related PDFs. These travel inside the PDF and inflate its size, often invisibly to the reader.

How to detect: a "paperclip" or "attachments" panel in your PDF viewer. The file is much larger than its page count would suggest.

How to fix: strip attachments in the source application before exporting, or use a PDF editor that can remove embedded files.

6. Verbose metadata and document properties

Metadata (title, author, subject, keywords, custom XMP fields) is usually small, but some applications write large XMP packets, embedded thumbnails, or full document histories. This is rarely the main cause, but it is easy to clean up.

How to detect: check File → Properties in your PDF viewer. A surprisingly large XMP block or embedded thumbnail is a sign.

How to fix: use theEdit Metadata tool to clear the fields you do not need, then compress.

7. Unoptimized object structure

PDFs written by older exporters store each small object (page descriptions, font descriptors, dictionaries) separately with its own overhead. PDF 1.5 introduced object streams that pack many small objects into one compressed stream. A document saved without them carries structural fat that a simple re-save can trim.

How to detect: you cannot easily from the outside, but this is the cause when a text-only PDF from an older tool is larger than expected and shrinks dramatically on a single compress pass.

How to fix: runCompress PDF. Re-saving with object streams is lossless and invisible to the reader — it is the safest lever and a good first move for any large PDF.

Which cause is it? A quick triage

1. Can you select text? If no, the PDF is scanned — re-scan at a lower DPI (cause #3).
2. Is it mostly images? If yes, downsample from the source (cause #1).
3. Is it text-only but still large? Run Compress PDF. If it shrinks a lot, it was unoptimized structure (cause #7). If it barely moves, suspect embedded fonts (cause #2) or metadata (cause #6).
4. Still stuck? ReadPDF Compression Explained and the file size guide for the deeper mechanics.