Image quality and file size: a practical guide
File size is driven by pixel dimensions, format, and compression settings — not by how large an image looks on screen. Understanding those levers lets you shrink files without visibly hurting quality, and tells you when a smaller file simply is not worth the artifacts.
Match the format and settings to the picture: JPEG or WebP for photographs, PNG or lossless WebP for graphics and transparency. Resize to the pixels you actually display, tune the quality setting, and keep a pristine original so you never re-compress an already-degraded file.
Pixel dimensions versus display size
An image's data is a grid of pixels — for example 4000 by 3000. That pixel count, not the size the image appears on a page, is what mostly determines file size. Shrinking a 4000-pixel-wide photo to the 800 pixels a card actually shows removes roughly fifteen out of every sixteen pixels, which is often the single biggest saving you can make.
Displaying a large image in a small box does not save bandwidth: the browser still downloads every pixel and then scales it down. If a layout never shows an image larger than 800 pixels wide, exporting it at 800 pixels wide is both smaller and sharper than shipping a 4000-pixel file the browser has to resample.
Resolution labels like 'DPI' or 'PPI' describe print size, not the on-screen file. For web use, think in pixels: decide the largest size the image will be shown at, then export at that pixel width.
What the quality setting actually does
For lossy formats such as JPEG and lossy WebP, a quality value controls how aggressively the encoder discards detail during compression. Higher quality keeps more information and produces a larger, cleaner file; lower quality discards more and produces a smaller file with visible artifacts such as blocking around edges and halos near high-contrast lines.
Quality is not a linear percentage of visible fidelity. On typical photographs, high settings look almost identical to the source while still saving space, and the visible cost of compression only grows quickly at the lower end of the scale. The practical method is to lower quality step by step and stop at the first setting where you can see a problem, then move back up one notch.
The three levers that change file size
| Lever | Effect on size | Effect on quality |
|---|---|---|
| Pixel dimensions | Large — fewer pixels means far less data | Only matters below the size you display at |
| Format choice | Large — lossy formats beat lossless on photos | Depends on content; graphics need lossless |
| Quality setting | Moderate — lower quality, smaller file | Artifacts appear at the low end of the scale |
Metadata: hidden weight and privacy
Photographs often carry metadata such as EXIF camera settings, timestamps, GPS coordinates, and color profiles. This data adds bytes and can leak private details — for example the exact place a photo was taken — when the file is shared publicly.
Removing metadata reduces both file size and privacy exposure. PixMovo strips metadata by default during image conversion, so exported files do not carry hidden location or device information.
Transparency changes your options
If an image needs a transparent background — a logo, an icon, a UI element — the format must support an alpha channel. PNG and WebP do; JPEG does not. Trying to save a transparent image as JPEG silently fills the background with a solid color.
For flat graphics with transparency, PNG is lossless and predictable. WebP can store the same transparency while often producing a smaller file, which is why it is worth testing WebP when transparency and size both matter.
Repeat encoding compounds loss
- Every save of a lossy JPEG or lossy WebP re-applies compression, so editing and re-exporting the same file repeatedly accumulates artifacts — an effect often called generational loss.
- Lossless formats such as PNG and lossless WebP reproduce the same pixels on every save, so a working master stays intact no matter how many times you open and export it.
- Keep a pristine original or a lossless master, and export a fresh lossy file only when you need the small final version.
- Converting a compressed JPEG to PNG afterwards cannot restore detail the JPEG already discarded; it only stores the degraded pixels losslessly.
Choosing between JPEG, WebP, and PNG
Reach for JPEG when you need a photograph that opens everywhere and file size matters more than pixel-perfect edges. It is the safest universal choice for continuous-tone images.
Reach for WebP when you want smaller photographs or graphics with transparency and your audience uses modern browsers; it supports both lossy and lossless modes. Reach for PNG when you need guaranteed lossless graphics, sharp text, or transparency with the broadest compatibility. PixMovo converts among JPEG, PNG, and WebP and strips metadata by default.
Worked example: shrinking a hero photo for the web
- Resize to the display width
Decide the largest width the image is ever shown at — say 1600 pixels for a full-width hero — and export at that pixel width rather than the camera's native 4000 pixels.
- Convert to JPEG or WebP in PixMovo
Upload the resized photo and choose JPEG for universal support or WebP for a smaller file. PixMovo strips metadata as it converts.
- Tune quality until artifacts appear
If your workflow lets you lower quality, step it down and watch edges and gradients. Stop at the first setting where blocking or banding shows, then move back up one notch.
- Compare against the original
View the export beside the source at full size. If the smaller file looks clean, ship it; if not, raise the pixel width or quality rather than accepting visible damage.
Use an image converter
Frequently asked questions
- Why is my image file so large even though it looks small on the page?
- File size follows the pixel dimensions of the image, not how big it appears in the layout. A 4000-pixel photo shown in an 800-pixel box still downloads all 4000 pixels. Resize to the width you actually display to cut the size.
- Does lowering the quality setting always ruin the image?
- No. Quality is not linear. On photographs, high settings look almost identical to the source while still saving space; visible artifacts appear mainly at the low end of the scale. Lower it gradually and stop before problems show.
- Should I use JPEG, WebP, or PNG?
- Use JPEG for photographs that must open everywhere, WebP for smaller photos or transparency on modern browsers, and PNG for lossless graphics, sharp text, and transparency with the broadest compatibility.
- Will converting a JPEG to PNG improve its quality?
- No. JPEG discards detail permanently. Saving it as PNG afterwards only preserves the already-degraded pixels losslessly; it cannot recover what JPEG removed.
- Does removing metadata make the file smaller?
- Yes, modestly, and it also protects privacy by dropping EXIF location and device data. PixMovo strips metadata by default during conversion.