A PDF linked from a web page is a performance asset like any other. An oversized PDF slows the download, eats mobile data, and frustrates users. Here is how to make web-ready PDFs.
PDFs are often treated as finished documents rather than web assets, but when a PDF is linked from a page and opened in a browser, it is a performance asset like any image or script. A 40 MB PDF on a mobile connection is a slow, frustrating experience, and many visitors will abandon the download before it finishes. Optimising PDFs for the web is the same discipline as optimising images: reduce the bytes without hurting the experience.
What makes a PDF large
A PDF is a container holding text, fonts, vector graphics, raster images, and metadata. Almost always, the raster images dominate the file size. A single page scanned at 600 DPI in full colour can exceed 8 MB; a ten-page scan can pass 80 MB. Embedded fonts add up when a document uses several families with full character sets. Duplicated objects — a logo re-embedded on every page instead of referenced once — inflate the file invisibly. Metadata, annotations, and revision history persist long after they are useful.
Step 1: Remove what is not needed
Before compressing, cut content. Delete pages nobody will read — appendices, blank separators, duplicated covers — with Organize & Delete PDF Pages or extract only the relevant range with Split PDF. Removing pages is the highest-quality compression there is, because the pages you keep are untouched. A thirty-page document trimmed to the eight pages someone actually needs is already a 70% reduction.
Step 2: Right-size the images
If images dominate, deal with them directly. Ask what resolution the document actually needs: 150 DPI is ample for screen reading, 200–300 DPI for office printing, and anything above 300 DPI is invisible at normal viewing distance and pure file weight. If you control the source images, compress them before assembling the PDF with the Image Compressor, then rebuild with Image to PDF. Starting from optimised inputs always beats compressing a bloated output.
Step 3: Run a structural optimisation pass
A lossless optimisation rewrites the file structure more efficiently: removing unused objects, de-duplicating repeated resources, compressing object streams, and stripping redundant metadata. Nothing visual changes. Run the document through the PDF Compressor, which rebuilds and re-optimises the structure entirely in your browser. For PDFs from word processors or design software, this alone frequently solves the problem.
Step 4: Consider lossy compression for scans
For scanned documents, lossy compression is where the dramatic reductions come from. Re-encoding the page images at a lower quality or resolution can take a 30 MB scan to 2 MB. At moderate settings the difference is invisible on screen and only apparent when zooming past 200% or printing at high resolution. For a document meant to be read on screen, that is a fine trade.
Step 5: Verify
Open the optimised file, view every page at 100%, and check small text, thin lines, tables, and logos — the first things to degrade. If the document will be printed, print one representative page. Confirm the file size meets your target. A web-hosted PDF should generally be well under 5 MB so mobile visitors on slow connections are not stranded.
Serving PDFs on the web
- Aim for under 5 MB for web-hosted documents; under 2 MB for mobile-first audiences.
- Link the PDF directly rather than embedding it in a heavy viewer that adds its own payload.
- Set the right Content-Type and support byte-range requests so viewers can stream large PDFs.
- Compress once from the highest-quality original; re-compressing compounds artefacts.
- Keep archival or signed originals untouched and optimise only the copy you publish.
A web-ready PDF is one a visitor can actually open. The same habits that make images fast — right-sizing, compressing, removing the unnecessary — make PDFs fast too, and every byte you remove is a byte that does not have to travel over a mobile network.