Why Converting PNG to JPG Makes It Bigger

You convert a PNG to JPG expecting a smaller file, because JPG has a reputation as the "compressed" format. Instead the JPG comes out bigger — sometimes noticeably bigger. This isn't a bug in your converter and it isn't rare; it's a frequent question on forums like the Adobe Community, and it happens because the "JPG is always smaller" assumption is only true for one specific kind of image: photos. JPG's compression is built around continuous-tone photographic detail — smooth gradients and subtle color variation. PNG's lossless compression, on the other hand, is genuinely efficient at flat colors and sharp edges. If you convert an image that's mostly flat color and hard edges — a screenshot, a logo, a UI graphic — into JPG, you're handing JPG's algorithm exactly the kind of content it's bad at compressing. The result: a bigger file, sometimes with visible artifacts added on top, for no benefit at all.

JPG's compression advantage comes from exploiting how continuous-tone images behave — a sky with a gradual color shift, skin tones, foliage, shadows that fade gradually. In that kind of content there's a lot of visual information a human eye doesn't easily notice, and JPG's algorithm throws that information away, then compresses what's left efficiently. That's where the size savings come from — it's not that JPG is a generically smarter format, it's that it specializes in one type of visual redundancy. Feed it a photo and it excels. Feed it something without that kind of subtle gradient content — flat colors, sharp text, hard edges between colors — and JPG has almost nothing to exploit. Worse, it still has to represent every hard edge somehow, and encoding a crisp boundary without visible blockiness can genuinely take more data than PNG's simpler, lossless approach needed for the same edge. JPG isn't bad at compression in general — it's specialized, and sharp edges are outside its specialty.

PNG uses lossless compression — it doesn't approximate or discard anything, and it reconstructs the image pixel-for-pixel exactly. That sounds like it should make PNG files larger across the board, but lossless compression is very good at one common pattern: long runs of identical or near-identical pixel values. Screenshots, app UI, logos, vector-style illustrations, icons, and text-heavy graphics are full of exactly that — big flat regions of a single color, bounded by sharp, clean edges. PNG's compression collapses those flat regions down efficiently, without approximating anything. So for this category of image, PNG isn't the "safe but bloated" option and JPG isn't the "efficient" one — it's the reverse. PNG is often already close to as small as the image can reasonably get, because there's very little redundant, hard-to-compress subtlety in the file for any algorithm to strip out. Converting that PNG to JPG doesn't remove waste — there wasn't much to remove — it just adds lossy artifacts on top of a file that gets bigger, not smaller.

For real photographs — camera photos, scanned images with continuous tone, anything with lots of subtle color gradients — converting to JPG should shrink the file, often substantially. That's the scenario JPG was built for. If you convert a genuine photo to JPG and the file doesn't get smaller, something else is going on rather than the format choice being wrong. The two most common culprits: the JPG quality setting is still set very high (near 100), which keeps most of the detail JPG would otherwise discard, or the file is carrying metadata bloat — EXIF data, embedded color profiles, thumbnail previews — that adds size independent of the actual pixel compression. Both are fixable without changing format: lowering the quality setting a modest amount usually recovers most of the expected savings, and stripping unnecessary metadata removes size that has nothing to do with image content at all. If a photo isn't shrinking on conversion, check the settings before concluding JPG doesn't work.

If the image is a screenshot, a logo, an illustration, an icon, or anything else with flat colors and sharp text or edges, PNG is very likely already the smaller, more efficient file — and converting it to JPG is likely to make things worse on two counts, not one. First, the file can get bigger, for the reasons above. Second, JPG's lossy compression introduces visible artifacts around every sharp edge and every block of text — soft haloing, blockiness, color smearing — that a lossless PNG never had. You lose quality and gain file size at the same time, which is the opposite of the trade-off format conversion is supposed to offer. There's no version of this where converting a genuinely flat-color, sharp-edged image to JPG is the right move for file size. If a PNG screenshot or logo feels too big, the fix is to compress the PNG itself, not to change its format.

The instinct to "convert to JPG to make it smaller" treats format conversion as a size-reduction tool, but it only works in one direction — photographic content. The actual fix depends on what the image is, not on swapping the file extension. If the file is too big, the right first step is usually a compress pass on the image in its current format — pixeltools' [Compress Image](/compress) tool handles both PNG and JPG and picks an appropriate compression approach for the image you actually have, rather than forcing a lossy re-encode you might not need. If you genuinely need a different format — for compatibility, transparency requirements, or a platform that requires JPG — use the [Convert Image](/convert) tool for that conversion specifically, understanding that the size outcome depends on the image content described above, not on the tool itself. Format conversion and file-size reduction are two different jobs; treating them as the same one is where this confusion starts.

There's no universal rule that settles this for every image, because it depends on the actual content of that specific file — how much of it is smooth gradient versus flat color and sharp edges, and most real images are some mix of both. A product photo on a plain white background, a screenshot with an embedded photo inside it, an illustration with soft shading — each can land differently. The only reliable way to know which is smaller for a given image is to test both: compress the PNG and separately convert-and-compress a JPG version, then compare the actual resulting file sizes. It takes a few seconds per image and removes the guesswork entirely. Don't assume either direction based on general reputation — check the specific file. PixelTools' own compressor bears this out at the code level: your chosen quality value is passed straight to the browser's canvas encoder with no per-image content analysis deciding the outcome for you, which is exactly why testing the actual file is the only real answer.