BlogImage Compression

How to Compress Screenshots Without Making Text Hard to Read

Learn how to compress screenshots while keeping text readable with the right format, lossless settings, cropping, resizing rules, and quick checks before sharing.

June 23, 2026

Updated: June 23, 2026

To compress screenshots without making text hard to read, keep text-heavy captures in PNG or lossless WebP, crop extra whitespace before changing quality settings, and avoid downscaling unless your display size requires it. JPEG and aggressive lossy settings often blur letters because they are built for photographs, not sharp UI text. Start with a lossless pass, compare the result at 100% zoom, then use a high-quality lossy format only if the screenshot contains photos or gradients and the text still passes a side-by-side check.

Key Takeaways

Default to lossless WebP for screenshots with text — it often produces smaller files than PNG while preserving text edges exactly.
Avoid JPEG for text-heavy screenshots unless required by a legacy system; even high quality settings can produce halos around letters.
Crop before you compress. Removing empty toolbar or sidebar space shrinks files far more safely than lowering quality.
Never downscale as a first move. Resizing destroys sub-pixel anti-aliasing; crop first, resize only if the platform requires it.
Verify at 100% zoom on both a desktop and a phone before publishing.

Why Compressed Screenshots Go Blurry

Text in a screenshot is poorly suited to typical "photo" compressors. Three failure modes explain almost every unreadable result.

1. Chroma subsampling

JPEG and many lossy presets store color (chroma) at lower resolution than brightness (luma) — a common scheme is 4:2:0 subsampling. Photographs hide this well because real-world color transitions are gradual. Anti-aliased text, however, relies on precise sub-pixel color to define letter edges. Subsample those edges and you get colored fringes and halo effects around compressed code snippets or UI captures.

2. DCT block quantization

JPEG breaks the image into 8×8 pixel blocks and rounds off their frequency components. High-frequency content — the sharp transitions that are text — gets rounded most aggressively. Lower the quality slider and the rounding step grows, smearing edges into the surrounding background.

3. Downscaling

Modern displays render text with sub-pixel anti-aliasing: tiny gray pixels along letter strokes that your eye reads as smooth edges. Even modest downscaling blends those anti-aliasing pixels into whatever sits next to them, producing the "soft, slightly fuzzy" text people often blame on compression when the real culprit is a resize that happened first.

Best Screenshot Format by Use Case

Use case	Recommended format	Why	Watch out for
Sharing in email or office docs	PNG (optimized)	Widest compatibility	Large file size without an optimizer
Publishing on the web	Lossless WebP	Often smaller than PNG with identical quality	Older CMS or email clients may not support it
Terminal or dark-mode UI with few colors	PNG-8 / pngquant	Palette reduction works well on flat colors	pngquant is lossy; check text edges after
Screenshot mixed with gradients or photos	Lossy WebP, quality 90+	Good balance of size and sharpness	Below quality 90, text halos become visible
Large screenshots in modern browsers	AVIF, quality 80+	Very efficient for supported apps	Check support in your target apps and CMS
Legacy systems that require JPEG	JPEG quality 95, 4:4:4 chroma	Last resort only	Generation loss compounds; keep originals

WebP supports both lossless and lossy modes. For lossless, Google's cwebp documentation describes the -lossless flag and the -z compression level (0–9; higher is slower but can produce smaller output). For browser support status, MDN's image format guide and Can I Use are the reliable references.

Step-by-Step Workflow

Step 1 — Capture at native resolution

Don't let your screenshot tool pre-shrink the output. On macOS, ⌘⇧4 usually captures at native pixel dimensions. On Windows, the Snipping Tool keeps the display's DPI unless a workflow resizes afterward. If you use a browser extension, check its output format and quality settings — many default to JPEG at moderate quality.

Step 2 — Crop before compressing

Every pixel of empty toolbar or sidebar you crop out removes bytes without touching the text you care about. A full-screen capture trimmed down to the relevant panel can be dramatically smaller before you change a single quality setting.

Step 3 — Choose your format

For a quick browser-based workflow, LessMB can compress supported image formats (JPG, PNG, GIF, WebP, AVIF) locally in your browser — files stay on your device — which is useful for making screenshots smaller before emailing or uploading them. Still compare the output against the original when small text matters.

For CLI work, cwebp -lossless -z 9 in.png -o out.webp produces a lossless WebP where -z 9 uses the slowest preset and can generate smaller output than faster presets.

Step 4 — Optimize or palette-reduce

Run an optimizer on the output:

PNG: oxipng -o max --strip safe file.png — lossless re-encoding with no pixel changes; exact savings vary by image.
WebP lossless: -z 9 in the encode step handles this; no separate optimizer needed.
Screenshots with few colors: pngquant --quality=90-100 file.png — this is lossy palette quantization, not strictly lossless, so always check text edges in the output before publishing.

Step 5 — Resize only if required, and resize correctly

If your platform caps width at a specific size, resize before the final lossy encode, not after. Use a high-quality filter:

ImageMagick: convert in.png -filter Lanczos -resize 1200 out.png
Photoshop: Image Size → Resample → "Bicubic Sharper"
Sharp (Node.js): .resize({ width: 1200, kernel: 'lanczos3' })

Avoid nearest-neighbor and bilinear resampling — both will damage anti-aliasing on text.

How to Check If Text Is Still Readable

Before you publish, run these checks:

100% zoom side by side. Open the compressed file and the original at full zoom. Text edges should look identical — watch for halos or color fringes around letters. If you see them, raise quality or switch to lossless.
Phone test. View the image on a small screen. If small text is unreadable there, dimensions are too small — re-export larger rather than lowering compression further.
OCR test. Drop the compressed file into an OCR tool (macOS Live Text works). If recognition accuracy drops compared to the original, you have gone too far with lossy settings.
Pixel diff. A pixel-diff tool will highlight ringing halos around letters that your eye might miss at a glance.

Red flags to look for: halos or glowing edges around letters, colored fringes, small text that disappears or merges, OCR errors that weren't in the original.

Common Mistakes to Avoid

Re-saving a JPEG screenshot multiple times. Each lossy re-encode compounds the degradation. Always compress from the original capture.
Assuming plugins and build pipelines preserve your format. Some CMS plugins, CDNs, or image pipelines may convert PNGs to JPEG or WebP automatically; check your media settings and test generated output.
Using "compress for web" presets built for photos. They are tuned for lower quality settings that work well for photographs but destroy fine text edges.
Adding drop shadows or rounded corners after compression. Apply visual effects to the source PNG first, then compress the composite.
Trusting file size alone. A small file with unreadable text is worse than a larger one that does its job.

Practical Next Steps

Set your default screenshot format to PNG so captures stay lossless: on macOS, run defaults write com.apple.screencapture type png in Terminal.
Add an oxipng -o max step to any blog or documentation build pipeline that processes images.
When you need to quickly compress a screenshot before emailing or uploading it, drop it into a browser tool like LessMB for local compression — then compare the result against the original before sending.

FAQ

What is the best format to compress a screenshot with text?

Lossless WebP is usually the best balance — it preserves text edges exactly like PNG but produces files that are often significantly smaller. Use PNG when you need maximum compatibility, and only fall back to lossy WebP or JPEG at high quality settings when the screenshot is mostly photographic.

Why does JPEG make screenshot text look blurry?

JPEG uses 8×8 block DCT compression and chroma subsampling — many JPEG encoders default to a 4:2:0 scheme — which discards the high-frequency color detail that defines sharp text edges. This produces ringing halos around letters and a muddy look, even at moderate quality settings.

Can I compress a PNG screenshot without losing quality?

Yes. Tools like oxipng or OptiPNG re-encode the PNG with better filters and DEFLATE settings without changing a single pixel. In many cases you can reduce file size without any quality loss at all.

Is WebP better than PNG for screenshots?

For web publishing, lossless WebP is often a better choice because it can produce smaller files while keeping text edges identical to PNG. PNG remains the safer pick when you need broad compatibility — for example in older email clients or office documents.

Should I resize a screenshot to make it smaller?

Avoid resizing whenever possible. Downscaling is the single biggest cause of unreadable screenshot text. Crop empty whitespace instead — trimming a large capture to the relevant panel removes far more bytes than resizing, with no risk to the text you want to keep.

What quality setting should I use for screenshots with text?

For lossy formats, stay at quality 90 or above for text-heavy screenshots. Below that threshold, DCT block artifacts and chroma shifts start appearing around letter edges. If text legibility is the top priority, use lossless WebP or optimized PNG instead.

How do I check whether compressed screenshot text is still readable?

View the compressed file at 100% zoom side by side with the original. Look for halos or color fringes around letters. On a phone, check that small text is still legible. For an objective test, run both files through an OCR tool and compare recognition accuracy.

Does compressing a screenshot twice make it worse?

Lossless formats (PNG, WebP lossless) can be re-saved indefinitely with no quality loss. Lossy formats (JPEG, lossy WebP, AVIF) accumulate generation loss every time you re-encode — always compress from the original capture, not from an already-compressed copy.