How to Make Image Quality Worse on Purpose

Quick answer

The fastest workflow is simple: upload the image, lower JPEG quality, check the preview, then reduce output scale if the result still needs to be smaller. Use pixel size for blocky damage, blur for soft detail loss, and noise or contrast for a crunchy degraded texture.

If you want a practical result, do not move every slider at once. Start with one primary source of damage. For most images, JPEG compression should be the main change because it creates the familiar artifacts people associate with a lower quality image.

What makes image quality worse?

Image quality gets worse when visible detail is removed, distorted, or hidden. That can happen through compression, resizing, pixelation, blur, noise, or repeated exports. A clean photo can turn into a lower quality image when edges become rough, gradients become blocky, text becomes softer, and small details disappear.

There is a difference between accidental damage and intentional damage. Accidental damage usually looks like a mistake. Intentional degradation uses the same visual clues, but controls them so the image still serves a purpose: a joke, a small upload, an old-web style, a rough screenshot, or a low-quality meme.

Best settings to make image quality worse

Use these ranges as starting points. The right settings depend on the original image, whether text must stay readable, and whether the result should look subtly worse or obviously broken.

Method 1: Lower JPEG quality

JPEG quality is usually the best first control because it changes both the visual texture and the downloaded JPG size. A setting around 35 to 50 creates a light quality drop. Around 22 to 35, the image starts to look copied or re-uploaded. Below 15, the artifacts become harsh enough for extreme damage or meme-style images.

This method works especially well for photos, gradients, and screenshots with thin edges. If the result is still too clean after lowering JPEG quality, add a small amount of pixelation or noise instead of forcing the quality value all the way down.

Method 2: Reduce output scale

Output scale changes the exported dimensions before the image is saved. It is useful when the goal is a smaller file, a lower resolution look, or a rough copy that travels faster through slow uploads. Scale around 70% to 90% keeps the image readable. Scale around 45% to 70% makes the quality loss more visible.

If the image will be displayed large, do not reduce scale too far. A very small output can look worse than intended because the browser or social app may stretch it back up.

Method 3: Add pixelation

Pixelation makes quality worse by grouping details into visible square blocks. Small values create a mild low-resolution feel. Medium values work well for low-quality memes, avatars, and copied images. Large values are best when losing detail is the point.

Use pixelation carefully on screenshots. Text can become unreadable quickly, so keep pixel size low if the image contains labels, captions, code, menus, or interface details.

Method 4: Add blur and noise

Blur makes an image look softer, less sharp, or more like a bad camera photo. Noise makes the result rougher and more chaotic. They are useful finishing controls after compression and scale are already set.

Avoid heavy blur when you want a crunchy degraded image. Crunchy images need visible edges and artifacts. Too much blur removes those edges and can make the result look soft instead of intentionally damaged.

When to use each method

Each quality-loss control has a different job. Choose the method that matches the problem instead of treating every setting as the same kind of damage.

Controlled experiment: one setting at a time

These examples use the same source image, the same crop, and the same comparison frame. Only one setting changes per card, which makes it easier to compare JPEG quality, output scale, pixel size, blur, and noise without mixing the effects together.

Same source, one changed settingEach card uses the same screenshot source and crop. Only one control changes per result, shown in the same comparison frame.

Source

900x900 JPEG

Original size

68 KB

Baseline

No effect

Goal

Controlled degradation

Original68 KBBaseline referenceClean source image

JPEG quality 2519 KB73% savedJPEG quality 25Best for visible compression artifacts and smaller JPG output.

Output scale 60%32 KB53% savedOutput scale 60% - 540x540 exportBest for shrinking dimensions before export and lowering KB.

Pixel size 856 KB19% savedPixel size 8Best for blocky low-resolution damage and meme-style texture.

Blur 240 KB41% savedBlur 2Best for a soft, out-of-focus low-quality look.

Noise 3569 KB1% largerNoise 35Best for a rough, crunchy, re-uploaded texture.

Common mistakes

The most common mistake is using too many effects at once. If JPEG quality, pixel size, blur, noise, saturation, and contrast are all pushed high, the image may stop looking intentionally low quality and start looking broken. Make one change, preview it, then add the next layer only when it improves the result.

Another mistake is adding noise before reducing file size. Noise can make an image look worse, but it can also make JPEG compression less efficient. If the goal is smaller KB, use Reduce Image Size KB first, then add noise only when the file is already close to the target.

FAQ