October 1, 2026 · 6 min read

Why Is My Photo So Big? Megapixels, HEIC and Screenshots

A single phone photo can be 8 MB. Here's exactly where that size comes from — resolution, format, HEIC, PNG screenshots — and what to do about each.

You try to attach four photos and the form says you're over the limit. Each one is 7 MB and you have no idea why — they're just pictures of a sofa. Photo size isn't mysterious, though. It comes from three things multiplied together: how many pixels the file holds, how the format stores them, and how much the camera decided to keep.

Reason one: you have far more pixels than you need

A 12-megapixel photo is roughly 4000 by 3000 pixels. A phone screen shows around 1200 pixels across; a laptop screen, maybe 1600 to 2500. So when you look at that photo on any screen, you are seeing a shrunken version of it. The extra detail is stored, transmitted, and then thrown away at the moment of viewing.

This is why downscaling is the single most effective thing you can do: it removes data nobody was ever going to see, rather than degrading data people will.

Reason two: the format

• JPEG is lossy and tuned for photographs. It's the reason a detailed 12-megapixel image fits in 3 MB instead of 36 MB. Every time you re-save a JPEG it loses a little more, so edit from the original where you can.
• HEIC is what modern iPhones shoot by default. It typically stores a similar-looking image in roughly half the space of JPEG — genuinely better, but older software, some web forms and plenty of Windows tools still won't open it, so people convert it and the file grows again.
• PNG is lossless and built for graphics with flat colours and sharp edges — logos, diagrams, screenshots of text. On a photograph it is spectacularly wasteful: the same picture can be five to ten times a JPEG.
• WebP compresses more efficiently than JPEG for the web and supports transparency, which makes it a good target format when you don't need PNG's exactness.

Reason three: screenshots are PNGs

Phones and laptops save screenshots as PNG, which is the right default — a screenshot of an interface is mostly flat colour and crisp text, exactly what PNG is good at. But a screenshot of a photograph, a map, a video frame or a web page full of images is a photograph wearing a PNG wrapper, and that's where you get a 9 MB screenshot of a chart. Re-saving it as JPEG or WebP can cut it by 80 percent or more with no visible difference.

The exception: if the screenshot is mostly small text that someone has to read, keep it lossless or use a high JPEG quality. Text is where JPEG artefacts are most obvious.

What to do, per situation

Skrubly's compressor does the first three of those: it redraws the image through a canvas at a smaller size and re-encodes it, so you get a downscale and a re-encode in one pass. Because it runs in your browser, the photo never leaves your device. If re-encoding would make a file bigger — which happens with tiny or already-optimised images — it hands the original back instead.

A side effect worth knowing about

Part of a photo's size is metadata: the camera model, lens, exposure settings, the exact timestamp and, if location services were on, the GPS coordinates of where you stood. It's a small fraction of the bytes but a large fraction of the risk. Re-encoding through a canvas drops it as a by-product — which is handy, but if privacy is the actual goal, do it deliberately with a tool that tells you what it removed.