How to compress a pdf to under 1 mb
- Step 1Open the compressor and drop your PDF — Load the document into PDF Compress (Aggressive). It works best on image-heavy or scanned PDFs.
- Step 2Set your target size — Enter a target like 1024 KB (1 MB). The tool will search for the highest image quality that still lands under it. If you'd rather not target a size, just pick a quality level and it compresses at that fixed quality.
- Step 3Let it search quality and resolution — Each page is rendered once and re-encoded as a JPEG at trial qualities; if even the lowest quality at full resolution overshoots, the tool lowers the render resolution and tries again — bounded so it always finishes.
- Step 4Download and check the report — The result shows the achieved size and whether the target was met. Download the compressed PDF and upload it wherever the limit applies.
Which compressor should you use?
The right choice depends on whether your PDF is text or images, and whether you need selectable text afterwards.
| Your PDF is… | Use | Keeps selectable text? | Why |
|---|---|---|---|
| A scan / photos / slide export | Lossy (this tool) | No | Pages are images already; re-encoding as JPEG shrinks them dramatically and can hit a hard size target. |
| Text with a few images | Lossy if size is critical, else lossless | Lossy: no · Lossless: yes | Lossy hits a small target but flattens text to an image; lossless keeps text but shrinks less. |
| Mostly text (a report, contract) | Lossless compress | Yes | Lossy would needlessly turn crisp text into a fuzzy image for little gain. |
| Bloated with metadata only | Metadata scrubber | Yes | If the bloat is metadata, stripping it is lossless and keeps everything else intact. |
How the target-size search works
Why this hits a number that a fixed-quality slider can't reliably reach.
| Step | What happens |
|---|---|
| 1. Render once | Each page is rasterised to a bitmap at the current resolution (the expensive step, done once per resolution level). |
| 2. Search quality | The tool binary-searches the JPEG quality and re-encodes the bitmaps (cheap) to find the highest quality whose assembled PDF is ≤ your target. |
| 3. Drop resolution if needed | If even the lowest quality at full resolution is still too big, the render resolution steps down (e.g. 150→110→90→72 DPI) and the search repeats. |
| 4. Stop at the best fit | The first quality/resolution combination that lands at or under the target is used; the report confirms the size. |
Realistic size expectations
Approximate — actual results depend on page count and image content. A 10-page colour scan is the hard case; a 2-page document is easy.
| Source | Typical original | Achievable target | Quality impact |
|---|---|---|---|
| 2-page colour scan | 4–8 MB | Under 1 MB | Slight softening, fully legible |
| 10-page colour scan | 20–40 MB | 1–2 MB | Visible softening at 1 MB; readable |
| Slide deck exported to PDF | 5–15 MB | Under 1 MB | Photos soften; text-as-image still clear |
| Mostly-text report | 1–3 MB | Lossless is better here | Lossy gives little benefit and loses text |
Cookbook
Common targets and what to expect.
Hit a 1 MB upload limit
The classic case — a job portal rejects anything over 1 MB.
targetSizeKb: 1024 → tool searches quality (and resolution if needed) → output: 0.94 MB targetMet: true
Squeeze under 500 KB for an email-friendly attachment
Tighter target; expect more softening on a multi-page colour scan.
targetSizeKb: 500 → output: 0.48 MB targetMet: true (resolution stepped down once)
Just make it smaller, no hard target
Skip the target and set a quality level instead (0.4 = aggressive, 0.7 = balanced).
quality: 0.5 (no targetSizeKb) → output: ~40–60% of original, one pass, fastest
Edge cases and what actually happens
The PDF is mostly text, not images
use losslessLossy compression turns crisp vector text into a fuzzy image for little size gain. Use lossless compression — it keeps text selectable and is the right tool for text-heavy documents.
You need the text to stay selectable / searchable
trade-offThis tool flattens each page to an image, so text selection and search are lost. That's inherent to lossy PDF compression. If selection matters, the lossless compressor is the only option — but it can't hit aggressive size targets.
Even the lowest quality can't reach the target
resolution dropFor a very large multi-page colour scan and a very small target, quality alone may not be enough. The tool automatically lowers the render resolution and retries; the result is more softening, but it still aims to land under your target.
The target is larger than the original
no-op-ishIf your PDF is already under the target, compression isn't needed — you can upload it as-is. Running the tool anyway will still re-encode it (and may even grow a tiny text PDF, since text becomes an image).
A scanned PDF with OCR text layer
OCR lostIf your scan has an invisible OCR text layer, lossy compression discards it along with the rest of the page structure. Re-run OCR after compressing, or keep the original if the searchable text matters.
Very large PDF (hundreds of pages)
slow but worksEach page is rendered in your browser, so very long documents take time and memory. It still completes; just give the tab a moment. Splitting the PDF first can help.
Frequently asked questions
Can it really get my PDF under 1 MB?
Yes — that's what the target-size mode is for. Set 1024 KB and the tool searches the JPEG quality (and lowers the render resolution if needed) until the output is at or under 1 MB, then tells you the size it achieved. For an image-heavy PDF that started at several megabytes this almost always succeeds; the only cost is some image softening.
Why does the text look fuzzy or stop being selectable afterwards?
Because lossy compression works by turning each page into a JPEG image. That's what makes scans and photo-heavy PDFs shrink so much — but it also means crisp vector text becomes part of the image, so it's no longer selectable or searchable and looks softer at low quality. If you need to keep selectable text, use lossless compression instead.
When should I use lossless compression instead?
Use lossless compression for text-heavy documents (reports, contracts, ebooks) and any time you must keep text selectable. Use this lossy tool for scans, photos, and slide exports — image-heavy files where small size matters more than text selection.
Is my document uploaded to a server?
No. Rendering and re-encoding happen entirely in your browser. The PDF never leaves your device — important for the personal documents people usually need under 1 MB, like CVs, passports, and bank statements.
What target sizes can I ask for?
Any size in kilobytes — 1024 for 1 MB, 500 for half a megabyte, 2048 for 2 MB, and so on. Smaller targets mean more aggressive quality reduction (and, if needed, lower resolution). Very small targets on long colour scans will visibly soften the images.
Will it work on a multi-page document?
Yes. Every page is rendered and re-encoded, and the target applies to the whole file. Long documents take longer because each page is processed in the browser, but the size target is applied across all of them together.
Does compressing twice make it smaller?
Not usefully — re-compressing an already-lossy PDF mostly just degrades quality further for little size gain. Compress once to your target. If you didn't hit the size you wanted, lower the target (or quality) and compress the original again rather than the already-compressed file.
What's the difference between quality mode and target mode?
Quality mode applies one fixed JPEG quality to every page in a single fast pass — good when you just want "smaller". Target mode searches for the best quality (and resolution) that lands under a size you specify — use it when there's a hard limit like 1 MB you must meet.
Why did my mostly-text PDF barely shrink — or get bigger?
Lossy compression is the wrong tool for text PDFs: a page of crisp text doesn't compress well as a JPEG and can even grow. Text PDFs are already small and should use lossless compression. This tool shines on images, not text.
Does it strip metadata too?
Re-rendering the pages naturally drops the old document structure and most metadata. If your only goal is removing metadata from an otherwise fine PDF, the metadata scrubber does that losslessly without touching the images.
Is there a file-size limit on what I can compress?
It's bounded by your browser's memory since everything runs locally. Typical documents are no problem; very large (hundreds of pages, hundreds of MB) files may be slow or memory-heavy — splitting them first helps.
Will the compressed PDF open everywhere?
Yes. The output is a standard PDF with one image per page, which every PDF reader, browser, and upload portal handles. It's about as universally compatible as a PDF gets.
Privacy first
All PDF processing runs locally in your browser using PDF-lib and pdf.js. No file is ever uploaded — only metadata counters are saved for signed-in dashboard stats.