DocoPDF
Guide

How to Reduce PDF File Size

By DocoPDF Team · 2026-11-20 · 1290 words

To reduce a PDF's file size, re-encode its embedded photos at lower quality and smaller pixel dimensions — images are what usually make a PDF heavy — and drop what the file does not need. Compressors do this on your device with quality presets; when a hard limit exists, aim for a target size instead.

Why PDF files get big in the first place

Most of a PDF's weight sits in its embedded images. A page of plain text is compact, but a scan, a screenshot or a photographed receipt is a full image, often at a resolution far beyond what a screen needs. Across many pages, images dominate the file.

The file's structure adds overhead too: loose objects, cross-reference entries and leftover data. Metadata rides along, though removing it usually saves little on its own. The takeaway: to make a PDF smaller, re-encode the images.

How to reduce a PDF file size, step by step

Compress with quality presets

  1. Open Compress PDF in your browser. The tool runs entirely on your device; the PDF is never uploaded.
  2. Drop in your file — up to five at once.
  3. Pick a compression level. Light keeps photos nearly identical to the original. Recommended is the balanced default for email. Strong produces the smallest files, trading visible image quality only as far as needed. Custom reveals two controls: JPG photo quality and a maximum pixel size per photo.
  4. Consider the extras. Grayscale converts colour photos to grayscale, squeezing scans and text-heavy documents down further. Metadata removal drops the document info and XMP block, but saves little on its own.
  5. Read the result. An honest before-and-after size readout shows exactly what you gained. Photos that would not shrink are left untouched, so the result never gets bigger.

The core promise: only photos are re-encoded. Fonts and vector art are never touched, so selectable text stays selectable.

Hit an exact size limit

When the requirement is a hard number ("the portal wants under 100 KB"), guessing quality settings is the wrong game:

  1. Open Compress PDF to Size and drop in your file.
  2. Choose the limit you were given: under 100 KB, 200 KB, 500 KB, 1 MB, or a custom number between 30 KB and 10 MB.
  3. Let it work. The tool walks a quality ladder, gentle text-preserving passes first, and stops at the first attempt that fits.
  4. Read the outcome. If even the strongest pass cannot reach the target, you get the smallest possible file with a clear note that it was not met — never a fake success. A file that already fits comes back untouched.

What compression can and cannot do

Be honest about the tradeoff: re-encoding photos is lossy. Lower quality means slightly softer-looking photos; that is the price of a smaller file. Presets exist so you choose how far to go.

When a PDF barely shrinks

Two situations produce disappointing results, and no honest tool can fix either:

  • Text-only documents. A pure-text PDF is already compact: the size lives in the photos, and there is little to win.
  • Already-compressed files. Images exported as well-compressed JPEGs have little left to give. The Strong preset or Grayscale may still help scans, but at some point the floor is the floor.

There is a hard ceiling too: a many-page file with dense text has a minimum readable size.

Common mistakes to avoid

  • Expecting lossless magic. If a file's size comes from photos, shrinking it means re-encoding those photos — anything promising big savings with zero change is selling something.
  • Compressing a scan that needs OCR first. Compression never adds a text layer. If your scan must become searchable, run PDF OCR to place an invisible, selectable text layer behind the page images — recognition reads those images, so do it while they are sharp, then compress.
  • Re-compressing repeatedly. Running an already-compressed file through again stacks quality loss for little extra savings. Compress once, from the original.
  • Stripping metadata and expecting a big win. It cleans the file, but the size lives in the images.
  • Using the wrong tool. Web Optimize PDF rebuilds the file's container — packing objects, re-deflating streams, dropping unreferenced junk — without touching a single image pixel. Great for structure-heavy files; pointless when photos are the weight.

When you should not compress a PDF

  • Legal or archival originals. Keep the untouched original and compress a copy; the original stays as it was.
  • When a portal requires PDF/A. Long-term preservation formats have strict rules, and re-encoding images can break compliance — check the requirement first.
  • When the real problem is too many pages. If the file is heavy because it holds chapters nobody needs, Remove PDF Pages or Split PDF cuts real weight with zero quality loss — the only truly lossless way to shrink.
  • When the file will not open. A PDF that fails to open is a repair job, not a compression job; Repair PDF rebuilds the broken structure first.

Frequently asked questions

Why is my PDF still big after compressing?

Because the size was never in the images. A text-only PDF is already compact: its weight is fonts and character data, so re-encoding has nothing to work on. Images that were already well-compressed JPEGs also have little left to win. For scans, the Strong preset or Grayscale can still help; for dense text, the answer is fewer pages or a bigger limit.

Will the text stay sharp and selectable?

Yes. Only the embedded photos are re-encoded. The text layer, fonts and vector graphics are copied through untouched, so search, copy and screen readers keep working exactly as before. Text never pays the toll: images do, and only as far as the setting you chose demands.

How do I get a PDF under 100 KB?

Use the target-size tool and set the limit to 100 KB. It runs progressively stronger passes, image re-encoding first with text kept selectable, and only converts pages to images as a last resort. If a many-page, text-dense file cannot go that low, you get the smallest readable version, clearly marked. Your options then: a bigger target, fewer pages, or a split.

Does removing metadata make a PDF smaller?

Only a little. Metadata removal drops the document info block and the XMP data: it cleans the file and removes tool fingerprints, but the savings are usually small because the real weight in most files sits in the photos. Treat it as hygiene, not compression.

Is compressing a PDF the same as optimizing it for the web?

No. Compression re-encodes the photos inside the file to make it smaller. Web optimization rebuilds the container, packing loose objects, re-deflating streams and dropping unreferenced objects, without touching a single pixel. Neither pass makes a file truly linearized for fast web view; an honest tool says so rather than stamping a fake flag.

Do online PDF compressors upload my file?

Most online tools send your PDF to a server to do the work. DocoPDF's compressor runs entirely in your browser: the file is processed on your device and never uploaded. That matters for contracts, ID documents, medical records and anything you would rather not hand to another server.

Shrink your PDF now

Open Compress PDF, drop in your file, pick a level — Light, Recommended, Strong or Custom — and get an honest before-and-after readout in seconds, entirely on your device. Facing a hard upload limit? Head to Compress PDF to Size and name the number the portal demands.