Compress a PDF Under 2 MB on Android (Free)

To get a PDF under 2 MB, reduce the resolution and colour depth of the images inside it — not the "quality" slider applied to the whole file. Almost every oversized PDF is oversized for one reason: it is a stack of full-page scans at 300 DPI in colour, each page an 8.7-megapixel photograph, and dropping those pages to 150 DPI greyscale often takes a 12 MB document to somewhere between 800 KB and 1.5 MB with the text still readable.

On a phone, Convertlix for Android does that step on the handset: its Compress PDF tool shrinks a PDF by re-encoding the images inside it, and the file is opened through Android's own file picker and never leaves the device — no upload, no server-side copy, nothing to retain. The scope: Android 8.0 or later, sign-in required (there is no guest mode), files up to 25 MB per job on the free tier against a server-set monthly allowance of conversions shown in the app, and output in the phone's Downloads folder. Everything below applies whichever tool you use.

Why the portal rejected it, and why "it's only 8 pages" doesn't matter

You're not imagining this — 2 MB is a tight cap, and it is the one you meet on government forms, university applications, visa portals, court e-filing and job systems. (10 MB is the other common number; if that is your limit, Get a PDF Under 10 MB for a Visa or Job Portal leaves more room.) The cap is almost never about page count — it is about how the pages are stored.

Half the problem is free to fix; half costs you something:

• The fixable half: resolution, colour depth and image encoding. On a scanned document that is 60–90% of the bytes, and cutting it costs almost nothing you can see on a screen.
• The half no tool fixes: the information itself. A compressor cannot invent back detail it removed, and it cannot make 400 pages into 8. If a reviewer needs 300 DPI colour on every page to verify a stamp or a hologram, 2 MB may not fit, and the answer is a second upload or a rescan.

Which kind of PDF do you have? Try to select the text

Two kinds of PDF blow past a cap, for opposite reasons, and the test takes five seconds: try to select a line of text.

• The text selects — it is born-digital, exported from Word, Pages or Google Docs. Text is stored as text and shapes as vectors, so eight pages is typically 100–400 KB. If yours is huge anyway, blame one embedded object: a full font family, a 4,000-pixel logo, a pasted screenshot.
• Nothing selects — it is a scan, from a flatbed, a copy shop or a phone scanning app. Each page is one image; there is no text in the file at all. An A4 page at 300 DPI in colour is 2,480 × 3,508 pixels — 8.7 megapixels per page, roughly 26 MB before compression. Ten pages is the storage equivalent of ten photographs, which is why Why Is My Scanned PDF So Large? is the better question here.

Building the PDF from photos in the first place? Control the size at assembly: Turn Photos Into One PDF for an Upload (Android).

What actually shrinks a PDF, in order of effect

Order matters: resolution is multiplicative — halving DPI removes three quarters of the pixels — so it is the first lever. Quality sliders come last, because that is where legibility goes.

How to get a PDF under 2 MB

1. Work out which kind of PDF you have. Try to select a line of text. If it selects, skip to step 5 — vector text is untouched by image compression. If it does not, you have a scan, and steps 2–4 hold your savings.
2. Downsample the images to 150 DPI. This alone takes most scanned documents from 8–15 MB to 1–3 MB. 150 DPI is above what any screen displays and still legible on paper; go lower only if the document will never be printed.
3. Drop to greyscale when the colour carries no information. A scanned contract, transcript or utility bill loses nothing. Do not do it to a passport, ID card, cheque or anything with a coloured stamp a reviewer has to verify.
4. Try bilevel encoding on pure black text. CCITT Group 4 and JBIG2 exist for exactly this case and routinely land at 30–60 KB per page. They destroy anything with shading, so check every page before you send it.
5. Re-encode the file with a compressor. On Android, Convertlix's Compress PDF re-encodes the images inside the file; on a desktop, any compressor with image downsampling will do. There is no target-size dial, so treat it as one pass, not a slider.
6. Check the output size, then open every page. Read the number off your Downloads folder before you go near the upload form, then look at the pages — compression is the step most likely to ruin the page the reviewer cares about.

When you can't get there without losing something real

Be honest about these, because a rejected application costs more than a second attempt:

• A colour scan of an ID document under 2 MB degrades visibly past about 8–10 pages. Split it across two uploads if the portal allows, or rescan at 200 DPI rather than compressing a 600 DPI file twice.
• Compressing twice is worse than compressing once well. JPEG artefacts compound and never come back, so start from the original rather than the compressed copy.
• A 400-page PDF cannot be compressed into the cap. That one needs splitting; no image setting closes a gap that large.
• A 500 KB cap is a different problem, not a harder version of this one — at that size text legibility becomes something you manage deliberately. See Compress a PDF Under 500 KB Without Blurry Text.
• If the portal also demands PDF/A, the two goals conflict: PDF/A forbids some compression schemes and requires embedded fonts. Convert to PDF/A last, then re-check the size.

FAQ

Will compressing make the text blurry?

Only if the text is part of an image. In a born-digital PDF the text is vector data and stays sharp at any zoom, however hard the images are compressed. In a scan the letters are pixels, so downsampling softens their edges: at 150 DPI that is invisible on screen, and below roughly 100 DPI it is obvious.

Why is my PDF still over 2 MB after compressing it?

Usually because the compressor changed something other than the images. Font subsetting, metadata stripping and "optimize for web" together save a few percent on a scan, where 90% of the bytes are pixels. Check that image downsampling was actually applied; if the file is born-digital, hunt for one huge embedded object instead — a screenshot, a logo, a map.

Does OCR make a PDF smaller?

No. On its own it makes the file very slightly larger: OCR adds a text layer on top of the page image rather than replacing it. It is worth running because it makes the document searchable and selectable, but size reduction has to come from downsampling or re-encoding the scan.

Does Convertlix upload my file?

No. The file is opened through Android's own file picker, converted on the phone and never sent anywhere, and the app asks for no storage permission — its single runtime permission is the camera, used only for scanning pages. Its network calls carry identity, licence, quota, configuration and payment data, never a document: sign in once, switch on airplane mode, and the tools still work.

Is there a free way to do this without an app?

Yes. Ghostscript is free and open source, and if you are comfortable with a terminal it gives finer control over downsampling than a phone app does: gs -sDEVICE=pdfwrite -dPDFSETTINGS=/ebook -o out.pdf in.pdf. On a Mac, Preview's Reduce File Size filter is built in. If you still have the source document, re-exporting from LibreOffice at a lower image resolution beats compressing the PDF afterwards, because the oversized images are never created. Browser converters work too, but as of September 2026 the mainstream ones upload your file to their servers and meter the free tier — check their pricing page for the cap.

---

If the file is on your phone and you would rather not hand it to a website to find out whether it fits, re-encode it locally and check the size in Downloads before you upload.