How to reduce an image to a specific size
A kilobyte (KB) is a unit used to measure digital file size. One kilobyte equals 1,024 bytes, and a 100KB image is therefore roughly one hundred thousand bytes of data. That sounds like a lot, but a single photo taken on a modern phone can easily reach 3MB or more — around thirty times larger than the limit many websites allow.
Why so many websites ask for a specific size
File-size limits exist for practical reasons. Government portals, job application systems, university admissions forms and visa services often cap uploads at 100KB, 200KB or 500KB because they store millions of files and have to keep their infrastructure manageable. On the web, size also affects speed: every extra kilobyte has to travel to the visitor's device, and large images are one of the most common reasons a page feels slow.
Compressing a JPG
JPG is a lossy format, which means the encoder throws away some visual detail to save space. That is exactly why it compresses so well. Lowering the quality setting from 100 to around 70 often cuts the file size by more than half with almost no visible difference on a screen. Below roughly 50, artefacts start to appear — especially around sharp edges and text.
Compressing a PNG
PNG is lossless. It stores every pixel exactly, and it supports transparency. That precision is why screenshots, logos and graphics with hard edges look so clean as PNG — and also why PNG files are large. You cannot simply "lower the quality" of a PNG, because there is no quality setting to lower.
When WebP is the better option
WebP was designed for the modern web. At the same visual quality it is typically 25–35% smaller than JPG, and unlike JPG it supports transparency. It is an excellent choice for images you are publishing on a website you control. The trade-off is compatibility: some older forms and desktop applications still expect a JPG or PNG.
Why dimensions matter so much
File size scales with the number of pixels in the image. A 4000 × 3000 photo contains twelve million pixels. Reduce it to 1200 × 900 and you are left with roughly one million — about 8% of the original data. In many cases, resizing is the single most effective way to reach a small target.
How FITLY reaches your target
FITLY runs entirely in your browser. When you choose a target size, the tool decodes your image, re-encodes it at a trial quality level, and measures the resulting file. It then adjusts — first the compression quality, and if that is not enough, the dimensions — and repeats until the file lands at or just under your limit. Because the whole process uses the Canvas API on your own device, your image is never uploaded to a server.