How to Reduce Image File Size Without Losing Quality

Reducing image file size is one of the highest-leverage optimisations available to any web developer, content creator, or business user. Done correctly, the result is visually indistinguishable from the original while being dramatically smaller. Done wrong, the result is obviously degraded.

This guide covers what actually works.

Understanding Image Compression

All image compression falls into one of two categories:

Lossless compression removes redundant metadata and internal patterns without altering any pixel value. The decompressed image is bit-for-bit identical to the original. PNG uses lossless compression.

Lossy compression permanently discards image data, trading quality for size. JPEG is the canonical lossy format. The "quality" or "compression" slider in any JPG tool controls how much data is permanently discarded.

The critical insight: you rarely need a lossless copy of a photograph. A JPG at quality 85 is visually indistinguishable from quality 100 to virtually all observers under normal conditions — but can be 60–70% smaller. The data being discarded maps to high-frequency detail that the human visual system does not prioritise.

Choosing the Right Format

Choosing the wrong format is the single largest avoidable source of unnecessarily large image files.

The Most Common Mistake

Saving a photograph as PNG "to preserve quality." PNG is lossless, so it stores every pixel of every compression artefact already present in the original, producing a file 3–10× larger than an equivalent JPG with no visible quality benefit.

The 4 Techniques That Reliably Work

1. Resize to Display Dimensions First

Compression only removes data within pixels. Resizing removes entire pixels. If you upload a 4000 × 3000 photograph to display at 800 × 600, you are storing 25× more pixels than the display will ever use.

Resize before compressing, always. Use our Resize Image tool to hit the exact dimensions the platform or design requires.

2. Set JPG Quality to 80–85

Quality 100 preserves every JPG artefact at maximum fidelity — it does not produce a "lossless" result; it just stores a very high-quality lossy result. Quality 80–85 is the engineering sweet spot:

• Typically 60–70% smaller than quality 100
• Artefacts are present but imperceptible under normal viewing conditions
• Below quality 75, ringing artefacts become noticeable around high-contrast edges

3. Strip Metadata

A typical JPEG from a smartphone embeds: GPS coordinates, camera model, lens specification, timestamp, shooting mode, thumbnail image, and colour profile. This metadata is between 20 KB and 150 KB per image — zero visual value, pure overhead for web use.

Our Compress Image tool strips all non-essential metadata automatically.

4. Convert to WebP for the Web

WebP (developed by Google, 2010) delivers:

• 25–34% smaller than JPG at equivalent visual quality
• 26% smaller than PNG for lossless images
• Full transparency support (unlike JPG)
• Animation support (unlike JPG)
• 97%+ global browser support as of 2024

If you are optimising images for a website, converting to WebP is the highest-return single change available. Use our WebP Converter.

Quality Settings Reference

WebP vs JPG: A Practical Comparison

At quality 85, a typical 1080 × 1080 photograph:

WebP delivers equivalent visual quality to JPG at roughly 30% smaller size — with no tradeoffs for web use.

Quick Decision Flowchart

Is the image a photograph?
  YES → Use JPG (Q85) or WebP (Q85)
  NO  → Does it have transparency?
          YES → Use PNG or WebP
          NO  → Does it contain text or sharp lines?
                  YES → Use PNG or SVG
                  NO  → Use JPG or WebP

Is the image for a website?
  YES → Convert to WebP for best results

Is the file still too large after compression?
  → Resize to display dimensions first, then compress