How to Reduce a PDF to Under 1MB (Without Making It Unreadable)

A visa portal caps photos at 100KB. An insurance claim form rejects anything over 2MB. A job application system has a 1MB limit on "supporting documents." These limits are everywhere, and a standard PDF export from Word, Acrobat, or a scanner routinely produces files four to ten times larger. Here is how to bridge the gap without making the document unreadable at the other end.

Why PDFs are larger than you expect

A PDF's file size is the sum of several components, and understanding which ones are large tells you which approach will work:

• Embedded images — the biggest contributor in most documents. A scanned PDF or a report with photographs can have images totalling tens of megabytes embedded at full resolution.
• Fonts — PDFs embed the fonts used in the document so they render identically on any device. A document using several uncommon fonts can carry a few hundred kilobytes of font data.
• Metadata — document properties (author, creation date, edit history, software used) are usually small but worth removing for both privacy and size.
• Annotations and form data — interactive forms, comments, and digital signatures add overhead.
• Unused objects — PDFs accumulate debris as they are edited: deleted pages may leave orphaned objects, form fields that were removed may have residual data structures.
• Version history — incrementally-saved PDFs stack changes rather than rewriting the file, which can significantly bloat size.

The right approach depends on what is making it large

Before compressing blindly, look at what kind of PDF you have:

Text-heavy document (report, contract, letter) — these are typically under 500KB already. If yours is large, the culprit is probably embedded fonts or metadata. Metadata removal and object cleanup (both achievable in a browser-based tool) usually bring these under any reasonable limit.

Scanned document (scan of paper) — these are essentially image files wrapped in a PDF container. The images are the entire size. A 10-page A4 scan at 300 DPI can easily be 10–30MB. Getting this under 1MB requires image recompression, which means some quality loss is unavoidable.

Mixed document (text with embedded photos) — the images drive size, but the text portions can be preserved perfectly. The right approach is to recompress only the images, leaving text and vector content intact.

Exported from Word / PowerPoint — these often contain images that were not optimised before export. Recompressing or reducing image DPI typically yields large savings.

Method 1: Browser-based metadata and object cleanup (no upload needed)

For text-heavy PDFs and documents where the source content is well-optimised, removing metadata and unused objects can achieve meaningful size reductions without any quality loss.

keptlocal's Compress PDF tool does this in your browser: it strips document metadata (author, creation software, modification dates) and removes orphaned objects and unused resources. No upload, no account.

What to expect: text-only PDFs often see 10–30% reduction. Image-heavy PDFs see less, because the images themselves are not recompressed — only the surrounding overhead is removed.

Method 2: Reduce resolution before exporting

If you still have the source file (Word document, InDesign file, presentation), the best time to address PDF size is before you export. Modern versions of Microsoft Word, LibreOffice, and Google Docs all have export settings that let you specify image quality.

In Microsoft Word: File → Save As → PDF → Options → Image Quality. Choose "Standard (publishing online and printing)" rather than "Best for printing." The difference in file size for a photo-heavy document can be 3–5×, with no visible quality loss for screen viewing.

In Google Docs: File → Download → PDF Document. Google's PDF export is reasonably well-optimised by default.

Method 3: Server-side compression for scanned documents

Scanned PDFs require re-encoding the embedded images at lower quality or lower resolution. This is where browser-based tools hit a wall — recompressing JPEG data inside a PDF properly requires a tool like Ghostscript (used by most professional PDF compressors), which runs server-side.

For scanned documents that need to be under 1MB:

• Adobe Acrobat (desktop) — File → Save As Other → Reduced Size PDF. Choose the compatibility version (Acrobat 9 or later is fine for most cases). For heavy scans, use Tools → Optimize PDF for more granular control over image resolution.
• Smallpdf or PDF24 — both offer server-side compression that uses Ghostscript under the hood. They upload your file. For non-sensitive documents, this is a pragmatic choice. For sensitive documents, consider the privacy trade-off.
• Ghostscript (command line) — free, open-source, runs locally: gs -sDEVICE=pdfwrite -dCompatibilityLevel=1.4 -dPDFSETTINGS=/ebook -o output.pdf input.pdf. The /ebook preset targets 150 DPI and typically reduces scanned PDFs by 60–80%. Use /screen for 72 DPI (aggressive) or /printer for 300 DPI (conservative).

Target sizes and what to expect from each approach

When quality loss is acceptable and when it is not

Compressing images inside a PDF degrades quality. How much matters depends on the use case:

Forms, applications, basic correspondence — moderate compression is fine. The recipient needs to read text and see basic images clearly. 150 DPI JPG compression at 75–85% quality is invisible at normal screen viewing sizes and on A4 printing.

Technical diagrams and engineering drawings — quality matters more here. Line drawings can become visibly blurry at aggressive compression settings. Use conservative compression or, if possible, keep text and vector elements as vectors (use PDF/A export from CAD software rather than printing to PDF).

Legal or contractual documents — the content must be legible and unambiguous. Keep a full-quality original. The compressed version sent for review or submission is a copy; the executed original should be preserved at full fidelity.

Medical imaging embedded in a PDF — do not compress these. Clinical imaging has specific requirements; a compressed JPG of a scan is not suitable for clinical use. For administrative documents (referral letters with embedded images), moderate compression is usually fine.

A quick checklist for hitting 1MB

1. Start with browser-based metadata removal — fast, lossless, covers text-heavy PDFs
2. If the source document exists, re-export from the source application with image quality set to "Standard" rather than "Best"
3. If the PDF is a scan and must be under 1MB, use Ghostscript's /ebook preset (150 DPI, ~70% JPEG quality)
4. Test readability at your target file size before submitting — open the compressed version at 100% zoom and read a paragraph or examine an image
5. Keep the original; the compressed version is for submission only

What "under 1MB" actually looks like for text documents

A 1MB text PDF is large by the standards of text-only documents. A typical A4 page of body text runs about 10–30KB in a well-exported PDF. A 30-page report with no images should comfortably fit under 500KB. If a text-only document is larger than 1MB, the most common cause is embedded fonts that were not subset (the full font file was embedded rather than only the characters used). PDF editors and Ghostscript can subset fonts during optimisation.

The 1MB limits that cause problems in practice are almost always imposed on documents containing images — passport-style photos, scanned signatures, certificates, and multi-page scans. For those, the approaches above give you the tools to hit any reasonable target.