Converting an image can feel like a no-brainer: pick a format, click a button, download the result. But if your source is a transparent PNG, the choice you make changes more than file size. JPEG doesn't support transparency, so transparent pixels have to turn into some visible color. PNG and WebP can keep them.
That difference matters for logos, cut-out product images, icons, and screenshots that you expect to sit on different backgrounds. A conversion that looks fine on a white preview can leave a noticeable white rectangle when you drop it onto a colored page. For a photograph, though, it may not matter much at all.
This guide walks through what happens to transparency when you convert PNG to JPEG or WebP, using two measured browser-based conversions. It also covers how to choose a format, what the quality settings actually do, what privacy claims mean (and don't mean), and a practical way to verify the output rather than trusting a file-size percentage or preview alone.
What happens to transparent pixels when you convert a PNG to JPEG?
JPEG doesn't have an alpha channel - that's the part of an image format that records how transparent each pixel is. So when you convert a transparent PNG to JPEG, the output can't keep those transparent pixels. Instead, the converter has to merge (composite) the image onto a solid background. In the browser converter used for the examples below, that background is white.
This isn't the same as "JPEG just making transparency less precise." With JPEG, there's no transparency in the output at all. A pixel that was fully transparent in the PNG becomes an opaque pixel with a color. Partially transparent edges get blended against the chosen background too, which can produce a light or dark fringe if you later place the image on a different background color.
A pixel-level conversion test
To make the behavior concrete, a 200 × 200 PNG was created with a transparent right half: 20,000 of its 40,000 pixels had zero alpha. Before conversion, the transparent corner read 0,0,0,0; an opaque corner read 6,6,14,255.
After converting that file to JPEG, the result was 3.6 KiB. The conversion panel reported a 200 × 200 image. When the output pixels were checked, none were transparent. The corner that had been transparent was now white: 255,255,255. That lines up with the converter's warning that transparent areas become white.
The same PNG was then converted to WebP. The 2.5 KiB output retained all 20,000 transparent pixels. In this test, transparency was preserved instead of flattened.
What this means for your image
If you convert a transparent image to JPEG, don't assume it will behave the way it looks on a preview. White might be fine if you're putting it on a document with a white background, but it can stand out as a visible box on a dark interface or a colored slide. Those soft, semi-transparent edges also inherit whatever background they were blended against.
If the image needs to stay transparent, don't pick JPEG. Choose PNG or WebP, then check the result in the actual place you'll use it. A white preview can make a flattened image look deceptively similar to one that still has real transparency.
When should you choose PNG, JPEG, or WebP?
There isn't one "best" format for everything. The right choice depends on what your image needs to do after conversion: keep transparency, preserve details, work with your target software, or stay within a file-size limit. If one of those constraints matters more than the others, that's usually your deciding factor.
| Format | Transparency | Common fit | Main trade-off |
|---|---|---|---|
| PNG | Yes | Logos, interface assets, screenshots and graphics that need clean edges | Can be larger than a lossy format for photographs |
| JPEG | No | Photographs and images intended for a solid background | Transparency is flattened; compression can introduce artifacts |
| WebP | Yes | Web images where transparency and smaller files are useful | Check that the software or workflow receiving the file supports it |
Choose PNG when the image has to stay dependable
PNG is a good default for transparent artwork, sharp text, icons, and assets you might edit again later. It keeps transparency and avoids the JPEG "background becomes visible" problem. It may not be the smallest option for every type of image - especially photos - but predictability often beats squeezing out a few extra kilobytes when the asset is part of a real workflow.
Choose JPEG for photographs without transparency
JPEG is usually a reasonable pick for a photograph with a solid background. In that case, flattening isn't really a loss - you might even benefit from a smaller file for sharing or publishing. It's still lossy, so the quality setting affects how much compression you can see. If you'll edit the image again later, keep the original - re-saving JPEGs repeatedly can compound quality loss.
Choose WebP when you need transparency and a smaller output
WebP can keep transparency while giving you different size/quality trade-offs than PNG. In the 200 × 200 test above, the WebP output was 2.5 KiB and kept the alpha pixels. That's specific to that test image, though - it's not a guarantee that WebP will always be smaller or look better in every situation.
A simple rule: use JPEG for photos that don't need transparency, and use PNG or WebP when transparent pixels matter. Then test the output where it will actually be used - your destination app, CMS, or browser. A format that "works" in theory only matters if the next step of your workflow accepts it.
How can you convert and verify an image in the browser?
A browser-based converter can change formats without sending the file to some separate conversion service. In the Image Studio example, the Convert tab accepts JPEG, PNG, and WebP files, supports multiple files, and sets limits of 25 MiB and 25 megapixels per image. The conversion panel offers PNG, JPEG, and WebP output, plus a quality slider from 10 to 100 that starts at 86.
The basic flow is straightforward, but a reliable workflow still includes checking the source, choosing the output format based on how the image will be used, and inspecting the saved result. This is especially important when transparency, client delivery requirements, or strict file-size limits are involved.
A practical conversion checklist
- Keep the original. Work from a copy, and make sure the source file is available if you need to redo the conversion. The tool described here creates a new copy and leaves the original untouched.
- Check whether transparency is present. If it's a logo, cut-out, or graphic, assume transparency might matter until you verify it.
- Choose the output for the job. Use PNG or WebP to keep transparency. Use JPEG only when a solid background is fine.
- Set quality deliberately. If the tool provides a quality setting, start with the default and adjust only after comparing results. A slider value isn't a universal measure of "looks good."
- Inspect the downloaded file. Confirm the extension, dimensions, appearance, and whether transparency survived - inside the software or page where you'll use it.
- Compare before and after sizes. Use the actual file sizes, but don't assume that a big percentage drop will repeat for your next image.
Look beyond the preview
The test converter showed the output format, dimensions, and file size. It also showed a percentage reduction. Those are useful details, but they don't answer the most important question for transparent images: did the alpha information survive?
If alpha matters, inspect the output against a contrasting background or use an image editor that can show transparency. If the result looks like a white rectangle, that's usually the transparent parts being flattened - not a preview glitch. And if you're testing a tool, checking the pixel alpha or inspecting the file in another application gives you stronger evidence than the conversion summary alone.
What do file size and privacy claims tell you?
Smaller files can be genuinely helpful, but one conversion isn't a benchmark for every image. In one test, a 640 × 480 photographic PNG of 713.8 KiB became an 18.2 KiB JPEG. The panel reported it was 97% smaller. The output kept the original dimensions, and the downloaded bytes began with the JPEG/JFIF marker. The conversion took 0.2 seconds in that run.
That drop happened for that specific file - but the input was a synthetic gradient with grain. Photos, screenshots, illustrations, and already-compressed images behave differently under compression. Treat these results as what happened to that example, not a promise that your image will also shrink by 97%.
Read quality settings as controls, not guarantees
In the browser converter, the quality slider runs from 10 to 100 and starts at 86. The examples were produced using that default setting. Because those tests didn't compare multiple slider values, they don't prove what every setting will do to every image, or which values will look acceptable for your particular use.
If file size is a hard requirement, generate a few versions and compare them at the size they'll actually be displayed. Look for visible artifacts around edges, texture loss, and color shifts. Keep the highest-quality version that still meets the delivery limit rather than picking a lower number just because it produces a smaller file.
Treat in-browser processing as a privacy benefit, not a full guarantee
Processing "in the browser" can reduce exposure compared with uploading the file to a remote converter. That's helpful for sensitive material like a passport scan, medical record, or unpublished client work. In the Image Studio description, conversion is done in the browser, and the original source file isn't replaced by the converted copy.
Still, "in the browser" isn't the same thing as a complete security review of a website, its scripts, or the device you're using. If the file is highly sensitive, follow your organization's data-handling rules and confirm that the workflow behaves as described. And don't treat a privacy statement as a substitute for checking the tool and the context you're using it in.
Conclusion
Once you account for transparency, the choice is pretty clear: JPEG can't keep it, while PNG and WebP can. In the tested conversion, transparent PNG pixels became white in JPEG, while the WebP output kept the alpha channel. That's the practical difference you should look for before you publish or hand off a file.
Pick JPEG when a solid background is acceptable. Use PNG or WebP when transparency is part of the design. Then inspect the downloaded file, compare real sizes, and keep the original if you might need to redo the conversion. Those checks take a few minutes and prevent the most common mistake: realizing too late that the background isn't transparent anymore.