thecompressjpeg

Compression ratio explained

A 10:1 ratio sounds impressive and tells you almost nothing on its own. Here is how to calculate it, and why the percentage saved is the more useful number.

The arithmetic

Compression ratio
original size ÷ compressed size. 5 MB → 500 KB is 10:1.
Percentage saved
(original − compressed) ÷ original × 100. The same file is 90% saved.
Ratio 2:1
50% saved. Typical of lossless on a screenshot.
Ratio 10:1
90% saved. Typical of quality-80 JPEG on a camera photo.
Ratio 20:1
95% saved. Visible artefacts on most content.
Measured against what?
The number is meaningless unless you know what the original was.

Reading the number properly

Two ways of stating the same thing

Compression ratio is the original size divided by the compressed size, written as a ratio. A 5 MB photograph reduced to 500 KB has been compressed 10:1.

Percentage saved is the same measurement from the other direction: the same file is 90% smaller. Both describe one relationship, and converting between them is straightforward — a ratio of n:1 means (1 − 1/n) × 100 per cent saved.

Percentages are generally the more useful form, because they compress the interesting range into readable numbers. The difference between 10:1 and 20:1 sounds like doubling; as percentages it is 90% versus 95%, which correctly signals that you are chasing the last sliver.

Why a ratio alone tells you nothing

A tool claiming 10:1 has said nothing until you know what it started from.

Compressing an uncompressed 36 MB bitmap to 3.6 MB is 10:1 and genuinely unimpressive — almost any encoder achieves it, because the source was maximally wasteful. Compressing an already-optimised 200 KB JPEG to 20 KB is also 10:1, and would require destroying the image.

This is why headline figures like "up to 90% smaller" are close to meaningless. They are true of an unoptimised camera original and wildly untrue of a file that has already been compressed once. The honest number is the one measured on your own file, which is why the results table on every tool here reports per-file figures rather than an average.

The diminishing-returns curve

The relationship between quality and size is steep and then flat, and the ratio makes that vivid.

Quality 100 → 90
Roughly 2:1 for no perceptible change. The best value available anywhere.
Quality 90 → 80
Roughly another 1.5:1, still with nothing visible at normal size.
Quality 80 → 70
Much less again, and artefacts begin in smooth areas.
Quality 70 → 50
Small further gain, clearly visible cost.

Almost all of the available compression is in the first two steps. Past quality 75 you are trading visible quality for a few per cent, which is rarely a good trade.

Resolution beats ratio

Because file size scales with pixel count, resizing produces ratios that no quality setting can match — and does it without touching the encoder's quality at all.

Halving the width and height leaves a quarter of the pixels, which is roughly 4:1 before any lossy compression happens. Combine that with a quality-80 encode and total ratios of 30:1 or more are ordinary, with the image still looking correct at the size it is displayed.

If you are chasing a ratio, look at the dimensions first.

A worked example

A 4032 × 3024 photograph from a phone, 4.8 MB, going into a blog post in an 800-pixel column.

  1. Resize the longest edge to 1600 px. About 900 KB — a ratio of 5.3:1 on its own.
  2. Encode at quality 80. About 180 KB — a further 5:1.
  3. Strip EXIF. A few kilobytes less.

Final: 4.8 MB to roughly 175 KB. That is a ratio of about 27:1, or 96% saved — and it looks identical in the column where it is displayed. Neither step alone would have got close.

Try it yourself

Everything below runs in your browser, so you can test any of this on your own files without uploading them.

  • Compress JPEG

    Shrink JPG and JPEG photographs with a quality slider or an exact target size.

    Open the tool

  • Resize image

    By pixels or percentage, aspect ratio locked, across a whole batch.

    Open the tool

  • Bulk compressor

    One setting across a whole folder, processed in parallel, downloaded as a ZIP.

    Open the tool

Questions

Frequently asked questions

Every answer below is present in the page source, expanded, so it can be read without opening anything.

What is a good compression ratio?

For a camera photograph headed to the web, 10:1 to 30:1 including a resize is normal and looks perfect. For an already-compressed file, anything above about 1.5:1 probably costs visible quality. There is no universal target, because the ratio depends entirely on how wasteful the original was.

Why is my compression ratio so low?

Almost always because the file was already efficiently compressed. A JPEG that has been through one good encoder has little redundancy left. Other common causes: the image is a photograph in PNG at high resolution, or the content is highly detailed — foliage, crowds, texture — which legitimately needs more bytes.

Can the ratio be less than 1:1?

Yes, and it happens. Re-encoding an already-compressed file at a high quality can produce a larger file, because the encoder now has to spend bytes faithfully preserving the previous encoder's artefacts. The tools here report that as an increase rather than hiding it.

Do different formats change the ratio?

Substantially. At matched visual quality WebP is typically 25–35% smaller than JPEG, and AVIF smaller again. Converting format changes the achievable ratio more than most quality adjustments do.

Is a higher ratio always better?

No. Past a point the ratio only rises because the image is being destroyed. A 50:1 ratio on a photograph means visible blocking and banding. The number to optimise is the smallest file that still looks right, not the largest ratio.

How do I measure the ratio on my own file?

Divide the original file size by the compressed size — and the results table on every tool here shows both figures, plus the percentage saved, for each file individually and for the batch as a whole.

From the blog

Related reading

Longer, more practical guides from the blog.

  • Image Compression

    JPEG Quality Settings Explained: What the Number Actually Means

    Quality 80 is not 80% of anything. It is an index into a quantisation table. Measured on one photograph: dropping from 100 to 90 removes 77% of the file and 4.2 dB of fidelity; dropping from 40 to 30 removes 17% and costs almost as much.

    8 min read

  • Practical Guides

    How to Compress Images for a Website Without Wrecking Them

    Resize before you compress. Measured on one photograph: half the width at quality 80 produced a 16.4 KB file, while full width at quality 40 produced 20.6 KB and looked far worse. The order of operations matters more than the settings.

    8 min read

  • Practical Guides

    How to Reduce Image Size for Email Without Sending a Blurry Mess

    Attachments are base64-encoded in transit, which inflates them by about a third. A 20 MB limit is really a 15 MB limit. Resize to 1600 pixels on the long edge and encode at quality 80, and a photograph lands near 150 KB.

    9 min read