Editorial Policy
What we publish, how we verify it, how games are tested before release, and the specific practices we rule out.
Last updated 1 September 2026
Content standards
Everything we publish has to pass three tests before it goes live: is it accurate, is it useful to somebody, and is it ours. An article that fails any one of those does not get published, regardless of how well it might perform in search.
In practice this means we do not publish pages that exist only to hold a keyword, we do not publish near-duplicates of our own articles with the words rearranged, and we do not publish a thin page today with the intention of improving it later.
Originality
All text on this site is written for this site. We do not scrape competitors, we do not spin other people’s articles, and we do not republish descriptions supplied by anyone else. All games are built in-house; no third-party game is embedded, licensed or rebranded here.
All visual assets are original. Game art, thumbnails, icons and illustrations are generated from vector drawing commands written for this site. Audio is synthesised at runtime with the Web Audio API rather than sampled from anywhere.
How games are reviewed and tested
A game is not published until it passes the nine-point checklist set out on our about page, run manually on at least one desktop browser and one phone. Testing covers loading, starting, keyboard input, touch input, collection, collision, scoring, combo stepping, game over, restart, pause, tab-hidden pause, local score saving and reduced-motion behaviour.
Because we are the developer as well as the publisher, we do not describe our own game pages as reviews and we do not award ourselves ratings. A game page is documentation, not a verdict.
How articles are researched
Our guides draw primarily on direct experience of building and testing the games described. Where a guide makes a claim about a browser API, an accessibility requirement or a web performance metric, it links to primary documentation — typically MDN Web Docs, the W3C Web Accessibility Initiative, or the Chrome team’s web.dev — so a reader can check it.
We do not cite research we have not read. We do not attribute statistics to unnamed “studies”. If a useful figure does not exist, the article says so and explains the concept without one.
Use of AI assistance
AI tools may be used for drafting support, outlining and editing passes. Every published page is reviewed, corrected and signed off by a human editor who is responsible for its accuracy. We do not publish unreviewed generated text, and we do not publish generated content that makes factual claims we have not verified ourselves.
Updates and corrections
Every game page and every article carries a published date and a last-updated date, and both are real. When a game changes in a way that affects what a page describes — a control, a value, a difficulty tier — the page is updated and its date moves.
Corrections are handled under the fact-checking policy. A substantive factual correction is noted on the page rather than made silently.
Editorial independence
The site is designed to carry advertising. Advertising has no influence on what we write, and no advertiser has any say over the content of any page. Advert placements are chosen to sit between content blocks and never over a game canvas, a control, a score readout or a navigation element.
If we ever publish sponsored content or accept an affiliate arrangement, it will be labelled clearly on the page itself. At the time of writing we do neither.
What we will not do
- Invent player counts, download figures, ratings, reviews or rankings.
- Describe a local high score as a worldwide ranking.
- Claim credentials, employers or awards we cannot evidence.
- Use fake download buttons, fake play buttons, fake notifications or misleading countdowns.
- Publish content that implies the virtual gems in our games have real-world value.
- Mass-publish pages to chase search volume.
Publishing cadence
We publish when there is something worth publishing. Our working ceiling is two substantial articles a day, and we routinely publish far fewer. Quality gates come before volume targets: an article that is not ready simply waits.