Who writes these guides

Last updated: 19 October 2026 · 4 min read

Every guide on this site is written, tested and published by the same person — the one who also wrote the tools. There is no content team, no freelance writers, and no articles produced in bulk and reviewed afterwards. When a guide says a number was measured, that number came out of a tool on this site, on a real image, and you can reproduce it yourself in about the same time it took to write the sentence.

The thing that makes this site unusual

There is a conflict of interest here, and it is worth naming rather than hiding. I both write the reviews and build one of the tools being reviewed. Any site in that position can tilt the comparison, so here is exactly how I try not to, and where I think you should still be sceptical.

How the tests are run

• The measuring tool is the tool being tested. When a guide quotes compression figures, those figures come from SnapTools. That is convenient, but it means a bug in my code would produce confidently wrong numbers. Treat the figures as a starting point for your own decision, not as an independent lab result.
• Test images are ordinary photos. Real phone camera output, screenshots and stock images — not carefully selected worst cases. A compression ratio on a flat, low-detail image will look far better than the same setting on a night-time street scene.
• Where upload behaviour is claimed, it was observed. In the six-compressor shootout, the upload claims come from watching what the browser actually sent over the network, not from reading each service's marketing page.
• Dates matter. Every article carries the month it was tested. Compression libraries, browser codecs and competing tools change; a 2026 number that was true last year may not be now.

Where I am probably biased

• I built the tool, so I am not a neutral party.
• I read the source of my own browser code, but I can only infer what other services' code does by observing network traffic from the outside.
• What is verifiable without taking my word for it: whether a file leaves your device. Open your browser's network tab, drop a photo in, and look. That claim needs no trust in me at all.

What gets left out

No sponsored posts, no affiliate links, no guest submissions and no paid placements. The funding article sets out what advertising does and does not change about the site. Where a tool cannot do something — HEIC input, AVIF output — the guides say so plainly instead of pretending otherwise; see the HEIC explanation for why those gaps exist.

Corrections and contact

If a number here is wrong, or a tool behaves differently than described, that is a genuine error worth knowing about. Send a message with the article name and what you observed — corrections get made and the article gets re-dated. As a small project I cannot promise a fast reply, but I do read them.

Start with the work, not with me

If you came here for a tool rather than a biography, the tool list is the fastest route, and the guides section is where the testing write-ups live.