Drop your file into the tool above and set the target to 2MB — it re-encodes the images inside the PDF to hit that size while leaving any real text layer untouched. Whether that gets you a crisp result or a visibly softer one depends entirely on what kind of PDF you’re starting with.
The two kinds of “large PDF” are not the same problem
A PDF’s size comes from two very different sources, and knowing which one you have changes what to expect:
Scanned or photographed documents. Each page is effectively a photograph — captured at a scanner or phone-camera resolution, often 200-300 DPI, and saved with light or no compression by default. A 20-page scanned document at 300 DPI color can easily hit 30-60MB. Because the “content” is really just image data, a PDF compressor can shrink it dramatically — 80-95% reduction is normal — by re-encoding those page images with more efficient JPEG settings and often downsampling to a lower (but still readable) DPI, like 150.
Native/typed documents. A report exported from Word, Google Docs, or LaTeX stores text as font glyphs and vector drawing instructions, not pixels. This is extremely compact — a 100-page text-heavy report might be 1-3MB before any compression at all, because there’s no image data to shrink. If a document like this is already large, the size is usually coming from embedded high-resolution photos or charts, and compression will target those specifically rather than the surrounding text.
Why 2MB is a common ceiling
2MB shows up constantly on government portals, university and job application systems, and grant/HR document uploads. It’s small enough to keep server storage and page load times predictable at scale, but large enough for a typical multi-page scanned form at a reasonably readable resolution. It’s also comfortably under most email attachment limits (around 25MB), so a 2MB PDF travels easily by email too.
What you lose at different compression levels
For scanned pages, compression mainly reduces two things: JPEG quality of the page images (introducing the same kind of soft blocking as an over-compressed photo) and DPI (effective sharpness — text becomes slightly less crisp but stays legible down to about 150 DPI for most fonts; below 100 DPI, small text starts to blur noticeably). For native documents, compression has almost nothing to touch besides any embedded raster images, so text quality is unaffected regardless of the target size.
Checking your result before you send it
Always open the compressed PDF and zoom into the smallest text on a representative page, particularly for scanned documents or forms with fine print, before submitting it somewhere that matters (a legal filing, an application). It’s a five-second check that avoids resubmitting later.
Frequently asked questions
- Why is my scanned PDF 40MB but a 60-page report I typed is only 2MB?
- A scanned PDF is really a sequence of full-resolution photographs (usually 200-300 DPI, uncompressed or lightly compressed TIFF/JPEG) stitched into a PDF wrapper — every page is an image, so file size scales with page count and scan resolution directly. A typed report is mostly vector text: letters are drawn from font outlines and coordinates, which take almost no space regardless of how many pages there are. This is why compression tools can shrink a scanned PDF by 90%+ but barely touch a text-only PDF — there's no image data to re-encode.
- Will compressing damage the text or make it unreadable?
- If the PDF contains real text (you can select and copy it), compression that targets the embedded images won't touch the text layer at all — it stays crisp and searchable at any compression level. If the PDF is a scan with no text layer, the 'text' you see is pixels in an image, and heavy compression can make it blurry or introduce artifacts around letterforms, the same way JPEG compression affects a photo. Check whether you can select text in the original before compressing a scan aggressively.
- Why do email providers and forms cap attachments at 2MB or 25MB?
- 2MB is a common ceiling for web upload forms (government portals, university applications, HR systems) built to handle steady traffic without large storage costs. 25MB is the more common email attachment limit (Gmail and Outlook both cap around there), so a 2MB PDF has comfortable headroom to be forwarded as part of a longer email thread without hitting that ceiling.
- My PDF has embedded fonts — will those get stripped?
- A general PDF compressor that targets image data and internal structure won't touch embedded fonts, since fonts aren't image data. If a specific tool advertises font subsetting or removal as part of its compression, that's a separate, riskier optimization that can occasionally cause rendering issues on systems without the original font — worth checking the output PDF's appearance after compressing if that's a concern.
- Can I compress a password-protected PDF?
- Not with a browser-based tool that never sees your password if the PDF is encrypted at the file level — the file needs to be decrypted first (in whatever software applied the protection) before its internal structure can be re-processed. If your PDF just has a password to open it, remove that protection first, compress, then optionally re-apply it.