When PNG Compression Is Not Enough for Smaller Image Files

You compress a PNG, check its size, and barely anything changes. The image still looks great. It also still exceeds your upload limit. Before trying another compression setting, check what the image contains. A screenshot, a transparent logo and a holiday photograph have different needs. Getting a smaller file often starts with choosing the right format.

Why PNG Compression Can Reach Its Limit

PNG uses lossless compression. It stores image data efficiently while allowing the stored pixel values to be recovered exactly. The official PNG specification defines lossless image storage and support for an optional alpha channel, which carries transparency information.

That approach works especially nicely for graphics with repeated colours and predictable patterns. Think of a diagram with a white background, black labels and a few coloured shapes. A photograph contains much more variation. Skin, grass, fabric and shadows can change slightly from pixel to pixel.

A PNG encoder can search for a more efficient way to store that information. It cannot promise a dramatic reduction while preserving every pixel. If an image has already been compressed efficiently, another lossless pass may save very little. That is a normal limit, rather than evidence that the tool has failed.

Compression results also depend on the starting file. An inefficient export may contain plenty of room for savings. An already optimised image may contain almost none. There is no reliable percentage that applies to every PNG.

Check What a Compression Setting Actually Changes

Some tools reduce PNG size by changing its colours before saving it. A reduced palette can produce a smaller image, but it may also alter gradients, shadows or fine detail. The resulting file remains a PNG. The process has still changed the original pixels.

Check whether a setting promises lossless optimisation or permits visual changes. Both approaches can be useful. They serve different needs, particularly when you are preparing technical screenshots or artwork that must retain exact colours.

Keep PNG When Its Strengths Match the Image

A smaller file is only useful if it still does its job. A help article needs readable screenshots. A logo needs clean edges. A product cutout may need to sit naturally against several background colours.

PNG is often worth keeping for these assets:

• Interface screenshots: Small labels and fine borders benefit from crisp reproduction.
• Flat graphics: Charts, diagrams and simple illustrations often compress efficiently.
• Transparent artwork: Logos, icons and cutouts may need clear backgrounds and partially transparent edges.
• Images requiring exact pixels: Some review, comparison and production tasks cannot accept compression artefacts.

For example, converting a settings screenshot to JPG may reduce its size but create faint smudges around letters. Those marks become distracting when readers enlarge the image to follow instructions. Keeping PNG and cropping unused space may deliver a better result.

A simple badge can also remain smaller as PNG than as a visually acceptable JPG. Format choice depends on the content. Test a representative image before applying one export rule to an entire folder.

Photographic PNG Files Often Suit JPG Better

A photograph does not become a better fit for PNG because an editing app saved it that way. Camera images, portraits, interiors and landscape photographs usually contain continuous tones and detailed textures. JPG is designed to compress this kind of content using controlled loss of information.

That trade-off can produce a substantially smaller publishing copy. The saving depends on the photograph, its dimensions and your quality setting. A useful result keeps faces, textures and lighting convincing at the size people will actually view.

Consider a restaurant photograph exported as a large PNG. It has no transparent areas and will appear within a blog post. Keeping every source pixel may provide little visible benefit to readers. A resized JPG at a carefully chosen quality setting may be a better delivery file.

For an opaque photographic image, a PNG to JPG converter can create a candidate for comparison. Keep the source file, then judge the converted copy by its appearance and file size. Changing the filename extension alone does not convert the image.

Watch for Text and Graphics Inside Photographs

Some images combine a photograph with pricing, labels or interface elements. These need closer inspection. A quality setting that suits the background photo may leave small text looking rough.

If you control the website layout, keep headings and prices as HTML text where practical. That lets you compress the photograph separately and keeps the wording readable. If everything must remain inside one image, compare PNG and JPG at the final display size before choosing.

Transparency Needs a Deliberate Decision

JPG does not support transparency. Converting a transparent PNG therefore requires a solid background. Depending on the application, the export may ask you to choose one or apply a default colour.

Check this before converting. An editor's chequerboard preview usually represents transparency. A white preview may simply mean the application displays transparent areas against white. Neither appearance alone tells you what the exported JPG will contain.

If a product cutout will always sit on a white catalogue page, flattening it onto white may be acceptable. If that same image must appear on a dark banner or coloured card, a white rectangle will become visible. Retain a transparent asset for those uses.

Inspect Soft Edges Against the Intended Background

Transparency includes more than completely clear areas. Hair, glass, soft shadows and curved edges can contain partially transparent pixels. These pixels blend with whatever sits behind the image.

Flatten the image onto its intended background before exporting to JPG. Then inspect the edges. A pale outline may appear when artwork prepared against white is placed against a dark colour. Stronger compression will not repair that mismatch.

Keep the transparent original even if today's publishing copy uses a fixed background. A later redesign may need a different colour. If transparency remains necessary and PNG is too large, a format that supports transparency, including WebP, may be worth testing against the destination's requirements.

Reduce Unneeded Pixels Before Chasing Smaller Files

Image dimensions often offer a larger opportunity than another compression pass. A photograph exported directly from an editing project may be much wider than its publishing space requires.

Check the largest size at which the image will appear. Allow for sharper displays where appropriate. An image shown at 800 CSS pixels wide might use a 1,600-pixel-wide source for a display with twice the pixel density. That is a planning example, not a universal requirement.

Crop empty borders and unwanted scenery before resizing. Then export at dimensions that suit the layout. Reducing dimensions changes the image, so perform this work on a copy.

Changing a DPI value without changing pixel dimensions will not meaningfully shrink the image data for web use. Look at width and height in pixels. Those measurements determine how much visual information the file must store.

Compare Exports Using the Same Starting Point

Create your candidates from one preserved source. This makes comparisons fair and avoids accumulating damage from repeated JPG saves. Exporting a PNG from an existing JPG cannot restore detail that earlier compression removed.

For a photographic asset, compare a resized PNG with a JPG using the same dimensions and crop. Start with a moderate JPG quality setting, then adjust according to the result. Quality numbers differ between applications, so a value of 80 is not a consistent standard across tools.

Review the candidate at its intended display size, then inspect sensitive areas more closely:

• Faces and fine textures: Check eyes, hair, fabric and foliage for lost detail.
• Smooth gradients: Look for visible steps or patchy transitions in skies and shadows.
• Letters and hard edges: Check for smudges, ringing or block patterns.
• Backgrounds and colour: Confirm that flattened areas and overall colour still suit the destination.

Select the smallest candidate that meets the actual publishing need. A slightly larger file can be the better choice when it preserves readable instructions or a convincing product image.

Check the Published Result Before Calling It Finished

An export preview is only part of the workflow. Upload the chosen file and view it in its real position. Confirm that the publishing system accepts the format, displays the correct crop and serves the intended image.

Some platforms resize or recompress uploads. Their final output may differ from your local copy. Inspect the published result on a phone and a larger screen, especially where text or transparent edges matter.

If you manage the website, check whether browsers receive an appropriately sized version. A small thumbnail does not need to download the same large file used for a full-width banner. Responsive image delivery can reduce transferred data without forcing every use into one export.

For batches, sort assets by their needs before processing them. Photographs, screenshots and transparent graphics deserve different settings. Trial those settings on a few examples before converting the whole collection.

Give Each Image a Format That Fits

When PNG compression stops delivering useful savings, reconsider what you need to preserve. Keep PNG for transparency, crisp graphics and exact pixels. Test JPG for opaque photographs where a controlled quality trade-off makes sense.

Keep an original, remove unnecessary dimensions and compare real exports. Treat transparent areas as a design decision before flattening them. The best smaller file is one that still looks right, works in its destination and leaves you with a usable source for the next edit.