How to Compress a PDF Without Losing Quality
Published: May 23, 2026
Compressing a PDF reduces its file size by processing the images embedded in the document — resampling them at lower resolution and re-encoding them more efficiently. Love2PDF's Compress PDF tool uses Ghostscript on a secure server to achieve 60–80% size reduction on typical scanned documents. The tool is currently in development — join the waitlist and we will notify you the moment it goes live.
Why PDF compression is different from other PDF operations
Merging, splitting, rotating, and reordering PDFs only rearrange existing content. No images are processed. No encoding changes happen. These operations can run entirely in a browser using JavaScript because they are fundamentally structural — moving pages around without touching their content.
Compression is different. A PDF file can be large for two reasons. The first is embedded images — scanned documents, photos, and diagrams stored at high resolution inside the PDF. The second is inefficient encoding of other elements. Meaningful compression — the kind that reduces a 50MB scanned contract to 8MB — requires decoding those embedded images, resampling them at lower resolution using a process called downsampling, and re-encoding them using efficient compression algorithms. This image processing pipeline requires native software. The industry standard is Ghostscript — the same engine used by print shops, publishers, and enterprise document management systems.
What Ghostscript does when compressing a PDF
Ghostscript analyses every image embedded in your PDF. For each image, it determines the current resolution (typically 300 DPI for scanned documents) and downsamples it to the target resolution. Screen quality targets 72 DPI — the smallest output, ideal for documents viewed only on screen. Ebook quality targets 150 DPI — a balance of size and readability suitable for digital distribution. Print quality targets 300 DPI — preserves sharpness for documents that will be physically printed.
In addition to image downsampling, Ghostscript removes redundant embedded data, optimises font storage, and restructures the PDF for efficient reading. The combination of these operations produces dramatically smaller files without visible quality loss for the intended use.
Visit Love2PDF to explore the full set of tools available today.
How much will compression reduce your file size
The reduction depends entirely on what is inside your PDF. Image-heavy PDFs — scanned documents, presentation decks, photo portfolios — compress dramatically. A 50MB scanned contract typically compresses to 6–10MB at ebook quality. An 80MB presentation deck with embedded photos typically compresses to 12–18MB. Text-only PDFs — generated from word processors with no embedded images — compress very little because the text is already stored efficiently as characters, not as images.
Why you should not use browser-based tools for compression
Several browser-based tools claim to compress PDFs. In practice, they can only do a limited form of optimisation — removing metadata, compressing font tables slightly, restructuring the file. They cannot meaningfully downsample embedded images because image processing at this level requires Ghostscript or similar native software. If you run a 50MB scanned PDF through a browser-based compressor, you will typically get a 48MB output. The file is technically smaller but the difference is not practically useful. Love2PDF is honest about this — our browser-based tools cover everything that can genuinely be done without a server. Compression is the one operation that genuinely cannot.
How Love2PDF compresses PDFs securely
When the Compress PDF tool launches, your file will travel over an encrypted HTTPS connection to Love2PDF's processing server. Ghostscript processes the file entirely in server RAM — it is never written to disk. The compressed result is held in memory and a one-time download token is issued to your browser. The moment your download completes, the file is automatically deleted from server memory. There is no copy. There is no log of the file contents. The server processes the file and forgets it.
This is different from Love2PDF's browser-based tools, where files never leave your device at all. For compression, the server step is unavoidable — but the server step is designed to be as brief and private as possible.
Why your PDF is large in the first place
Before compressing, it helps to know why the file is large. PDFs grow large for three distinct reasons, and the fix is different for each.
The most common cause is scanned images. When a physical document is scanned, each page is stored as a high-resolution photograph inside the PDF — typically 300 to 600 DPI. A 20-page scanned contract at 300 DPI can easily reach 50MB. Compression resamples these images at lower resolution, which is where the dramatic size reductions come from.
The second cause is embedded vector graphics and fonts. PDFs from design software embed every font and graphic as a separate object. Font subsetting — keeping only the characters actually used — removes the excess. This is lossless and happens automatically.
The third cause is redundant metadata. Deleted pages, edit histories, hidden layers, and thumbnail caches accumulate inside a PDF over time. Stripping these adds no visible change and can reduce file size by 10-30% on its own.
Quick diagnostic: open your PDF properties and look at page count vs file size. A 20-page text-only PDF at 2MB barely compresses — the text is already stored efficiently as characters, not images. A 10-page scanned PDF at 40MB will compress dramatically — each page is a high-resolution photograph stored as an image.
Three compression quality levels — which to choose
Most PDF compressors offer quality levels with vague names like "high", "medium", and "low". These map to specific DPI targets that determine how aggressively embedded images are resampled. Here is what each level actually does.
| Level | Target DPI | Best for | Typical reduction |
|---|---|---|---|
| Screen | 72 DPI | Email attachments, web upload, on-screen reading only | 80–90% |
| Ebook | 150 DPI | Digital distribution, sharing, archiving documents you read but rarely print | 60–75% |
| 300 DPI | Documents you will physically print — preserves sharpness | 20–40% |
For most practical purposes — emailing a contract, uploading a report to a portal, sharing a presentation — ebook quality at 150 DPI is the right choice. The quality reduction is invisible on screen and in standard print. Screen quality is only appropriate when file size is the absolute priority and the document will never be printed.
How much will compression reduce your file size
The reduction depends entirely on the content of your PDF. Here are realistic expectations based on document type at ebook quality (150 DPI).
| Document type | Typical original size | After ebook compression | Notes |
|---|---|---|---|
| 20-page scanned contract | 40–60 MB | 5–9 MB | Black and white scan compresses most aggressively |
| 50-page presentation with photos | 80 MB | 12–18 MB | Colour images compress less than black and white |
| 10-page text-only Word export | 1–3 MB | 0.8–2.5 MB | Text-only PDFs barely compress — no images to resample |
| 100-page scanned textbook | 150–250 MB | 20–45 MB | Largest absolute reductions on high-page-count scans |
If your PDF is already small (under 2MB) or is a text-only export from a word processor, compression will have minimal effect. The largest gains are always on scanned documents — each page is a photograph, and photographs compress dramatically when resampled from 300 DPI to 150 DPI.
Why browser-based compression tools often do not work
If you have tried a browser-based PDF compressor and found the output barely smaller than the original, this is why: genuine PDF compression requires image resampling, and image resampling requires native software. It cannot run in a browser tab.
Browser-based tools can perform lossless optimisations — stripping unused metadata, subsetting fonts, deduplicating repeated objects. These are useful and fast. For text-heavy PDFs, lossless operations alone reduce size by 15–40%. But for image-heavy PDFs — the ones that are 40MB, 80MB, 100MB — the images are the problem, and lossless operations do not touch them.
Meaningful image resampling requires Ghostscript or an equivalent native library. Ghostscript reads the PDF, decodes every embedded image, resamples it to the target DPI using a high-quality algorithm, and re-encodes it. This pipeline needs to run as native code — it is computationally too intensive and requires libraries that are not available inside a browser's JavaScript sandbox.
This is why Love2PDF is honest about how Compress PDF works: it requires a server. For everything that genuinely runs in a browser — merging, splitting, rotating, converting to images — Love2PDF processes files locally with no upload. For compression, the server step is unavoidable. What we control is what happens on that server: your file is processed in memory, never written to disk, and deleted the moment your download completes.
While you wait — other tools that reduce PDF size
While the Compress PDF tool is in development, here are other approaches that can reduce PDF file size in specific situations. Split the document to extract only the pages you need — a 50-page PDF where you only need pages 1-10 is much smaller as a 10-page document. Convert specific pages to images using PDF to Image and share the images instead of the full PDF. Merge only the relevant sections using Merge PDF rather than sharing the entire source document.
Frequently asked questions
Q: Can I compress a PDF without losing quality?
A: It depends on what you mean by quality. Ghostscript's ebook quality setting (150 DPI) produces files that look identical on screen and in most print scenarios — the quality reduction is only visible under close inspection at high zoom. For documents that will only be viewed on screen or shared digitally, ebook quality compression produces no perceptible quality loss.
Q: Why is my PDF so large in the first place?
A: The most common cause is scanned images. When a physical document is scanned, each page is stored as a high-resolution image (typically 300–600 DPI) inside the PDF. A 20-page scanned document at 300 DPI can easily be 50–100MB. Compression resamples these images to a lower resolution that is still perfectly readable.
Q: When will Love2PDF's Compress PDF tool be available?
A: The server infrastructure is in development. Join the waitlist on the Compress PDF page and we will notify you the moment it goes live.
Q: Why is my PDF barely smaller after compression?
A: Text-only PDFs — generated from word processors like Microsoft Word or Google Docs — contain text stored as characters, not as images. These files have very little to compress. The dramatic reductions (60–80%) happen on scanned documents where each page is stored as a high-resolution photograph. If your PDF is a text export with no embedded images, lossless compression alone will reduce size by 10–20% at most.
Q: Can I compress a PDF without uploading it?
A: Not meaningfully, for image-heavy PDFs. Browser-based tools can strip metadata and optimise fonts — useful but limited. Resampling embedded images — the main source of compression — requires Ghostscript running on a server. Love2PDF's Compress PDF tool uses server processing but handles your file in memory and deletes it immediately after download. For all other operations, Love2PDF processes files locally with no upload required.
Join the Compress PDF waitlist — get notified the moment it launches.
Join Waitlist →