PDF compression sounds simple: make the file smaller. But anyone who has clicked "Maximum Compression" on an important document and watched the result turn into a pixelated mess knows it is not that straightforward.
The problem is not compression itself. The problem is that most guides and tools treat compression as a one-size-fits-all slider when the right approach depends entirely on what kind of document you have, where it will be viewed, and what matters most for your specific situation.
This guide explains how PDF compression actually works under the hood, what each compression setting does to your document, and how to choose the right approach for your specific needs. The goal is to give you enough understanding to make confident decisions rather than guessing.
• What makes PDFs large and what compression actually changes
• Compression level comparison with expected results
• Step-by-step instructions for compressing any PDF
• DPI settings explained for different use cases
• Common mistakes and how to avoid them
• When compression is not the right solution
• Privacy considerations for document processing
What Makes a PDF File Large?
A PDF is a container format that can hold text, fonts, raster images, vector graphics, annotations, form fields, metadata, and structural information. These components do not contribute equally to file size.
| Component | Typical Contribution to File Size | Notes |
|---|---|---|
| Raster images (photos, scans, screenshots) | 80% to 95% | By far the dominant factor in most PDFs |
| Embedded fonts | 2% to 10% | Depends on number of fonts and character sets |
| Vector graphics (logos, diagrams, charts) | 1% to 5% | Usually negligible in size |
| Text content | Less than 1% | Text is extremely space-efficient in PDF format |
| Metadata, annotations, structure | Less than 0.5% | Minimal impact on total size |
The implication is clear: when you compress a PDF, you are overwhelmingly compressing the images inside it. Text, vector art, and structure contribute almost nothing to the total file size. A PDF that contains no images at all is typically already as small as it can reasonably get.
What Happens During PDF Compression
PDF compression is not a single operation. It is a sequence of techniques applied to different parts of the document. Understanding each one helps you predict how different settings will affect your output.
Image Re-encoding
The compressor reads each image embedded in the PDF and re-saves it using different compression parameters. Images originally stored as high-quality JPEGs can be re-encoded at a lower quality setting. PNG images (which are lossless by nature) can be converted to JPEG, which discards some visual data but achieves a much smaller file size. The trade-off is between visual fidelity and file size. The compressor can also convert color images to grayscale or reduce color depth, which cuts size further.
Downsampling (Resolution Reduction)
If a PDF contains a 300 DPI image that will only ever be displayed on a screen, keeping all 300 DPI of detail is wasteful. Downsampling reduces the image resolution to a target DPI. For example, reducing a 300 DPI image to 150 DPI cuts the number of pixels by 75%. On a standard monitor, the difference between 150 DPI and 300 DPI is invisible because the screen itself cannot display that level of detail. On a printed page, the difference is obvious.
Metadata and Redundant Data Removal
PDFs accumulate metadata during creation: author names, software versions, document creation and modification dates, editing session logs, embedded thumbnails, and document previews. A compressor can strip metadata that is unnecessary for the end reader. This step has zero impact on document appearance and typically saves 1% to 3% of total file size.
Object Stream Optimization
Internally, a PDF stores content as a series of objects. Modern PDF compressors group multiple small objects into compressed streams (a feature introduced in PDF 1.5). This reduces the overhead of managing many individual objects. This optimization affects only the internal structure of the file, not the visible content. The result is a slightly smaller file with identical appearance.
Font Subsetting
When a PDF embeds a font, it normally includes the entire font file with all available characters. Font subsetting replaces the full font file with a version that contains only the characters actually used in the document. If a document uses 60 characters out of a font that contains thousands, the embedded font can shrink from several megabytes to tens of kilobytes. The reader sees identical text.
Compression Levels: What Each Setting Actually Means
The names vary by tool, but most PDF compressors offer four compression levels. Here is what each one actually does to your document.
| Level | Typical Size Reduction | JPEG Quality Equivalent | Downsample Target | Suitable For |
|---|---|---|---|---|
| Low | 10% to 30% | 85% to 95% | No downsampling or minimal | Documents destined for professional printing, presentation on large displays, or archival as masters |
| Medium | 30% to 60% | 70% to 85% | 150 to 200 DPI | General business documents, client proposals, reports shared via email or portal |
| High | 60% to 80% | 50% to 65% | 100 to 150 DPI | Email attachments with size limits, web uploads, internal team archives |
| Extreme | 80% to 95% | 20% to 40% | 72 to 100 DPI | Draft review, internal previews, documents for systems with strict size caps |
The JPEG quality percentages above are approximations. Different compression tools may use different internal scales, but the principle is consistent: lower quality produces smaller files with more visible artifacts.
Step-by-Step: How to Compress a PDF
The following process works with the PDFCraft Compress PDF tool, which runs entirely in your browser. No file upload is involved.
Step 1: Check the Starting File Size
Before compressing, note the current file size. On Windows, right-click the file and select Properties. On macOS, select Get Info. This baseline lets you measure the effectiveness of your compression.
Step 2: Determine Where the PDF Will Be Used
Different destinations have different requirements. A PDF destined for high-end printing needs different treatment than one that will be viewed on a mobile phone. Ask yourself: will this document be printed professionally, printed on an office printer, viewed on a desktop monitor, viewed on a mobile device, or archived for storage? Your answer determines the right compression settings.
Step 3: Choose a Compression Level
Use the comparison table above as a starting point. When uncertain, choose Medium. You can always compress further, but you cannot restore quality lost to overly aggressive compression.
Step 4: Set the Target DPI
DPI (dots per inch) determines how much image detail the compressed PDF retains. The appropriate DPI depends entirely on how the PDF will be displayed or printed.
| Destination | Recommended DPI | Reason |
|---|---|---|
| Professional printing | 300 | Standard print resolution; lower values produce visible dots |
| Office printing | 200 | Most office printers cannot benefit from detail above 200 DPI |
| Desktop screen viewing | 150 | Adequate for all standard monitors; further reduction may soften fine details |
| Mobile viewing | 100 | Phone and tablet displays have high pixel density but small physical size |
| Archival / long-term storage | 72 to 100 | Lowest acceptable quality; suitable only when regular viewing is not expected |
Step 5: Compress and Inspect the Result
Run the compression. Download the output. Compare file sizes. Then open the compressed PDF and inspect each page at 100% zoom. Focus on areas with photographs, gradients, logos, and fine text. If you notice blocky artifacts, color banding, or text distortion, increase the quality setting and try again.
Step 6: Rename the File to Distinguish It
Append a label to the filename: proposal-2026-compressed.pdf or portfolio-screen.pdf. This prevents confusion between the compressed and original versions and helps you avoid accidentally sending the wrong file.
Common Mistakes in PDF Compression
Applying Maximum Compression to Every Document
Maximum compression is designed for one purpose: making the file as small as physically possible. It achieves this by applying aggressive JPEG compression and heavy downsampling. For a document that someone else will read, maximum compression is almost always the wrong choice. The visual degradation signals carelessness. Most documents need Medium or High compression, not Maximum.
Overlooking DPI Settings
Many compression tools apply a default DPI setting that users never adjust. If the default is 150 DPI and the PDF is destined for print, the output will look noticeably softer than the original. Always verify the DPI setting matches your actual use case rather than relying on defaults.
Compressing an Already-Compressed PDF
When a compressed file is still too large, the natural instinct is to compress it again. This causes generational quality loss: each pass degrades the images further while producing smaller and smaller size reductions. The correct approach is to return to the original uncompressed PDF and choose more aggressive settings from the start.
Using Aggressive Compression on Scanned Text Documents
Scanned documents contain text that is already less sharp than digitally created text. Aggressive JPEG compression introduces artifacts around letter edges, making the text harder to read. For scanned documents, use a higher JPEG quality setting (80% or above) or lossless compression if available.
Assuming a Smaller File Is Always Better
File size is one factor among many. A 2 MB document that looks distorted will not serve its purpose. An 8 MB document that looks professional will. The right file size is the smallest size that still represents the document faithfully for its intended audience and purpose.
When Compression Is the Wrong Tool
Compression addresses image data, but file size problems sometimes have better solutions that do not involve quality trade-offs:
- Reduce the number of pages. If the PDF contains pages that can be removed, extracting only the relevant pages eliminates their entire file size with zero quality loss. This is always preferable to compressing a bloated document.
- Convert color pages to grayscale. Color images contain roughly three times the data of grayscale images. For documents where color serves no purpose (text-heavy scans, black-and-white forms), converting to grayscale before compression can reduce size by 60% to 80% with no loss of useful information.
- Resize oversized pages. A PDF created for large-format printing (A2, A1, or larger) contains proportionally more image data. Resizing the pages to a standard dimension before compressing often produces better results than compression alone.
Privacy Considerations for PDF Compression
PDF compression requires software to read and process every element of your document. When you use a web-based compression service, your file must be uploaded to the service provider's server. For many documents, this is acceptable. For documents containing personal information, financial data, legal content, medical records, or trade secrets, uploading to a third-party server introduces risk that may not be necessary.
The PDFCraft Compress PDF tool performs all processing locally in your browser using JavaScript. The file is read on your device, processed on your device, and delivered as a download from your device. No copy of the file is transmitted to any server at any point. When you close the browser tab, the file data is cleared from memory.
For organizations operating under data protection regulations such as HIPAA, GDPR, or the Privacy Act, local processing is not merely convenient but may be required for compliance when handling protected information. If this applies to your organization, verify that your chosen compression tool does not transmit file data externally before using it with regulated documents.
Frequently Asked Questions
Does compression affect OCR text in scanned PDFs?
Some compression processes can strip invisible OCR text layers. If your PDF has been OCR-processed to make text searchable, verify that your compression tool preserves these layers. The PDFCraft tool preserves all text layers, annotations, hyperlinks, and form fields during compression.
Can I compress a password-protected PDF?
Most browser-based compression tools require an unlocked PDF. If you own the document, use a PDF unlock tool to remove password protection before compressing. Re-apply protection afterward if needed.
Will compression remove my digital signature?
Compression modifies the file content, which invalidates most digital signatures. Apply compression before adding a digital signature. A compressed-and-signed document will have a valid signature; a signed-and-compressed one will not.
How small can a typical PDF get?
A text-only PDF with no images can often be compressed to 10 KB to 50 KB per 100 pages. An image-heavy PDF realistically reduces by 60% to 85% while maintaining good visual quality. Reductions beyond 90% typically introduce visible artifacts in photographs and gradients.
Why does my compressed PDF look fine on screen but pixelated when printed?
Screen resolution is approximately 100 pixels per inch. Print resolution ranges from 300 to 1200 dots per inch. An image that appears sharp on screen contains significantly less detail than a printer can reproduce. If print quality matters, preview the compressed PDF at 200% to 300% zoom to simulate printed output.
Does compression change the page dimensions?
No. Compression reduces the data within each page but does not alter the page dimensions. To change page size, use a dedicated PDF resizing tool.
Summary
PDF compression is a practical tool for reducing file size, but it works best when the settings match the document's actual use. The three factors that matter most are: choose a compression level appropriate for the document's purpose, set the DPI to match the destination (print versus screen), and always start from the original uncompressed file.
A compressed PDF that looks professional and arrives reliably is more valuable than a perfect-quality PDF that cannot be delivered. The goal is not the smallest possible file; it is the smallest file that still meets the document's requirements.
Try the PDFCraft Compress PDF tool to compress your documents locally in your browser.
Related Articles
- How to Merge PDF Files Online Free
- How to Edit PDF Files Online Free
- Add Watermarks to PDF for Protection and Branding