Initializing, please wait a moment

When to compress vs convert an image - decision guide


"Compress" and "convert" sound like they solve different problems - and they do - but most people reach for the wrong one. Compression reduces file size without changing format; conversion changes format and may or may not reduce size. Getting the decision right can take a 3 MB iPhone HEIC to a 120 KB JPG you can email, or in the opposite direction, save your source PNG from an invisible quality cliff. This guide covers the difference, the routing rules, and the right tool for each job.

30-second decision. Ask yourself three questions:

  1. Is the recipient's platform the problem? (HEIC on Windows, WebP in older email clients) → Convert.
  2. Is file size the problem, but everyone opens the current format fine?Compress.
  3. Both?Convert first, then compress. Converting typically changes size too, so compressing the already-converted file is where you squeeze the last bytes.

Concrete routing: HEIC/HEIF → HEIC to JPG. Oversized JPG/PNG → Compress Image. Large folder-of-photos to email → ZIP after compressing.


What compression actually does

Image compression reduces file size while keeping the format unchanged. There are two flavours and they behave very differently:

Lossy compression (JPG, WebP lossy, AVIF lossy) throws away pixel detail that the human eye is least likely to notice. A JPG at quality 85 looks indistinguishable from a JPG at quality 100 on most screens, but the file is roughly a third of the size. The catch: each save-and-reopen cycle re-encodes the image and loses more detail. Five edits of a JPG at quality 85 is visibly worse than one edit at quality 85 of the original source.

Lossless compression (PNG, WebP lossless, GIF) finds patterns in the pixel data and encodes them more efficiently without touching a single pixel value. The output is bit-for-bit identical after decoding. PNG saves about 30-60% on UI screenshots and flat-colour logos; it saves almost nothing on photographic gradients.

When someone says "compress my image" they usually mean "make this JPG smaller." Sometimes they mean "make this PNG as small as possible without breaking it." The difference between lossy and lossless matters for the right tool choice.


What conversion actually does

Conversion changes format. HEIC to JPG, PNG to WebP, SVG to PNG - each of these replaces the encoding scheme. Conversion almost always changes file size too, because different formats compress differently, but size is a side-effect; the main purpose is compatibility.

The clearest signal that you need conversion rather than compression: the file opens fine on your device but fails on the recipient's device. HEIC is the classic case - iPhones produce it by default, Windows Photos can't read it without a codec install, and most CMS upload widgets reject the extension outright. Compressing the HEIC wouldn't fix any of that; converting to JPG does. Similarly, pasting a copied SVG into a third-party tool that accepts only PNG needs an SVG to PNG converter, not compression.


Routing table: 10 common scenarios

Starting pointGoalActionTool
iPhone HEIC photoEmail to Windows userConvertHEIC to JPG
30 iPhone HEIC photosBatch share to recipientConvert all, then ZIPHEIC to JPGZIP
5 MB JPG from an SLR cameraWeb upload under 2 MBCompressCompress Image
2 MB PNG screenshotSmaller for a docCompress (re-encode with tighter PNG filter)Compress Image
SVG logo from a designerPaste into a tool that only accepts PNGConvertSVG to PNG
3 MB PNG photoSmaller web-ready fileConvert to JPG (lossy), then optionally compressCompress Image (format switch)
Old JPG photo saved at quality 95Tighter web fileCompress (re-save at quality 82)Compress Image
Photo with transparency (product cut-out)Web-ready smaller fileKeep PNG, compress; convert to WebP if audience allowsCompress Image
PNG of a scanned receiptArchive as PDFConvert to PDF via the HEIC-to-PDF or image-to-PDF pathHEIC to JPG (PDF output)
4K screenshot at 8 MBShare over SlackCompress first; convert to JPG only if it's a photo, not a UI grabCompress Image

"Compress" when the format is already correct

Use compression when the recipient opens your current format fine and only the size is wrong. Typical triggers:

  • Gmail rejects your 27 MB PDF attachment and you need to fit under 25 MB.
  • Your CMS upload caps at 2 MB and your JPG is 4 MB.
  • A forum wants a product photo under 1 MB.
  • Page-speed audit flags your hero image at 800 KB when the page weight budget is 200 KB.

For JPGs, re-encoding at a lower quality (typically 80-85) is the fast win. For PNGs, the equivalent is a smarter PNG encoder (pngcrush, zopflipng, oxipng) - the file size usually drops 10-30% with zero visual change. See our JPG vs PNG for web guide for the quality-vs-size sweet spots.


"Convert" when the format is the problem

Use conversion when the current format itself is the blocker. Typical triggers:

  • iPhone HEIC photos that Windows Photos shows as a blank placeholder. Fix: convert to JPG.
  • SVG source file but a downstream tool only accepts raster. Fix: convert to PNG.
  • PNG photograph at several megabytes where a JPG would be under 400 KB. Fix: convert to JPG.
  • Old WebP image that a marketing manager's legacy Outlook client doesn't render. Fix: convert to JPG.
  • Scanned receipts as individual PNGs that the accounting team wants as a single PDF. Fix: convert (and combine) to PDF.

If you're converting because size is also a problem, pick the destination format that serves both goals. HEIC → JPG is universal, not optimal for size. HEIC → WebP is better for size but fewer recipients can open it. Pick universal when in doubt; almost every "save bytes by converting" plan fails on the one recipient whose Outlook 2016 doesn't render WebP.


The both-at-once case: convert, then compress

Sometimes you have a large PNG photograph you want to ship as a JPG under 300 KB. That's a two-step job:

  1. Convert PNG → JPG. File drops from ~3 MB to roughly 600 KB at quality 85, because JPG encodes photos more efficiently than PNG.
  2. Compress the JPG by dropping quality further (75-80) or resizing dimensions if visual detail allows. Each step trims another 20-40%.

Don't try to do both at once with a single "compress PNG" operation. PNG compression alone can't touch lossy-encoding gains, and a bad tool will either produce an oversized PNG or silently rewrite your file as JPG without telling you.


Anti-patterns that waste time

Compressing a HEIC to fix a Windows compatibility problem. Compression doesn't change format. Windows still can't open the result. Convert to JPG instead.

Converting SVG → PNG to "make it smaller." SVG logos are usually already smaller than the PNG equivalent. Convert only when a downstream tool rejects SVG.

Re-saving a JPG at quality 100 "to preserve quality." Quality 100 JPG is barely distinguishable from lossless PNG in file size, and you've still lost detail from the previous save. If archival integrity matters, keep the source PNG or RAW file; derive JPGs from it on demand.

Converting PNG screenshot → JPG to save bytes. Works for photographs. Fails on UI text and line art - JPG compression artefacts around sharp edges look worse than a slightly-larger PNG.

Converting every image to WebP. WebP wins on the wire in modern browsers. It loses on the one recipient whose email client or export target doesn't support it. Use JPG/PNG as the canonical; WebP as an enhanced variant served by <picture> with a fallback.


Tool picks, for reference

  • Compress JPG, PNG, or WebP (format unchanged): Compress Image - in-browser, no upload, quality-controlled re-encode.
  • Convert HEIC → JPG / PNG / WebP / PDF: HEIC to JPG - iOS 18 support, batch, local processing.
  • Convert SVG → PNG or WebP: SVG to PNG - for raster-only downstream tools.
  • Convert PNG → SVG (trace): PNG to SVG - for logos and line art only; not photographs.
  • Batch share a folder of photos: ZIP after compress/convert.

FAQ

Does compressing an image reduce its quality?

Lossy compression (JPG, WebP lossy) does; lossless compression (PNG, WebP lossless) doesn't. A JPG at quality 85 is visually indistinguishable from the original for most use cases; below quality 70, compression artefacts become visible. For photos, a modern one-step lossy compress to quality 80-85 is the baseline. For screenshots and line art, keep lossless.

Which format is the smallest for the web in 2026?

For photos: AVIF < WebP (lossy) < JPG. For UI and flat colours: WebP (lossless) < PNG. AVIF is about 20-30% smaller than WebP at comparable quality, and WebP is about 25-30% smaller than JPG. The trade-off is browser and email-client support: universal JPG/PNG still wins where reach matters more than bytes.

Can I compress and convert in one step?

Yes, with the right tool. Compress Image can drop quality and re-encode to a different format in a single pass - upload a PNG, choose "JPG" as the output, and set quality 82, and you get the converted-and-compressed result without running two tools. Batch jobs are the one place single-pass helps; one-off conversions usually benefit from explicit conversion first, compression second, so you see the intermediate file size.

What does "re-encode" do to image quality?

Re-encoding means decoding the image back to raw pixels and encoding again with new settings. For lossless formats (PNG, WebP lossless), there is zero quality loss. For lossy formats (JPG, WebP lossy, AVIF), the new encoding introduces fresh compression artefacts even if the quality setting is the same or higher. If you re-save a quality-85 JPG at quality 85, the output is measurably worse than the input. Keep a source file (PNG, HEIC, RAW, TIFF) and derive lossy versions from it on demand instead of iterating on the JPG.

Back to the Image Converter hub for every converter tool, or the Image Tools hub for editors, compressors, and optimizers.

Why trust these tools

  • Ten-plus years of web tooling. The freetoolonline editorial team has shipped browser-based utilities since 2015. The goal has never changed: get you to a working output fast, without an install.
  • Truly in-browser - no upload. Every file-processing tool on this site runs in your browser through modern Web APIs (File, FileReader, Canvas, Web Audio, WebGL, Web Workers). Your photo, PDF, audio, or text never leaves your device.
  • No tracking during tool use. Analytics ends at the page view. The actual input you paste, drop, or capture is never sent to any server and never written to any log.
  • Open-source core components. The processing engines underneath (libheif, libde265, pdf-lib, terser, clean-css, ffmpeg.wasm, and others) are public and audit-able. We link to each one in its tool page's footer.
  • Free, with or without ads. All tools are fully functional without sign-up. The Disable Ads button in the header is always available if you need a distraction-free run.