We get asked roughly once a week how we decide what to publish. The answer is mostly boring: we play the game, we ask three questions, and if any answer is no, the game does not make the catalog.
Filter 1: Does it work?
Sounds basic. About a third of submissions fail this filter. Things we check: Does it load on a mid-range Android phone (we test on a Pixel 6a) over LTE? Does the audio work without a click-to-enable prompt that is never wired up? Do controls work on touch as well as keyboard? Does the game survive a tab refresh? Does it survive backgrounding the tab and returning to it? You would be surprised how many games released in 2025 fail one or more of these.
Filter 2: Is the gameplay genuinely interesting for at least five minutes?
This is where most submissions die. Many HTML5 games are 30-second loops with no meaningful variation. They are fine — they have a place — but they do not sustain a five-minute session, which means a player has nothing to bring back. We are looking for games that either introduce new mechanics over time, reward repeated play with skill development, or have meaningful difficulty that scales with the player.
Filter 3: Does it respect the player's time and attention?
No false-urgency timers. No buy lives or wait 4 hours mechanics. No social-pressure invitations. No video ad pre-rolls inside the game frame. We license the game itself; we do not accept games whose framework injects external ads or trackers we cannot audit. About one in ten games we would otherwise license fails this filter.
If a game passes all three, we move to the review phase.
Reviews are written by humans who actually played the game
Specifically, by an editor who played at least one full session — five minutes minimum, often longer for puzzle games. We take notes during play, not after. Notes capture: first impression, point at which the game became fun (or did not), point at which it stopped being fun, comparable games we would reach for instead, and whether we would return.
The star rating reflects editor opinion, not engagement metrics
Some sites rate by completion rate or session length. We rate based on a single question: would I tell a friend to try this? Five stars: yes, with enthusiasm. Four stars: yes, with caveats. Three stars: only if you are specifically into the genre. Two stars: probably not. One star: do not bother. We do not list any one- or two-star games — if it is that bad, we just decline to publish.
What we do not do
- We do not accept payment for placement.
- We do not mark sponsored content because there is no sponsored content.
- If a developer provides a game for review (no money changes hands), we say so in the review.
- We do not run network ads on the game frame itself.
- We do not track players inside games — any analytics happens at the catalog level.
What we would like to do better
Honestly: more games per month. Our pace is currently about two new titles a month. The bottleneck is editorial — playing, reviewing, writing. We would rather publish slowly and well than quickly and badly, but the gap between us and aggregators is partly a content-volume gap, and we know it.
If you are a developer with a game you think might fit, the submission email is in our Contact page. We read every submission. We accept fewer than one in five.