APPSINITYSpam policy
Spam on Appsinity is any content that wastes a player's attention or manufactures the appearance of trust. Both are treated as integrity violations against the store, not as marketing.
Open Appsinity StudioMetadata spam
Anything that games the discovery surfaces rather than describing the game.
| Practice | Why it is spam | Effect on you |
|---|---|---|
| Keyword-stuffed title | Search and cards show junk to players | Listing is a review risk |
| Irrelevant category or tags | Puts the game in front of the wrong audience | Bad traffic, worse reviews |
| Same game as two listings | Floods browse with one product | Duplicates are removed |
| Near-identical reskins | Manufactured catalogue breadth | Enforcement can group them |
| Description that is only links | No information, only destinations | Removal of the text |
| Screenshot gallery of another game | Bait and switch | Violation, not spam |
| Constant retitle and re-tag | Chases trending queries | Manual review |
Fabricated engagement
The Terms are direct: ratings and reviews must reflect genuine user experience and must not be fabricated, purchased, coordinated or manipulated. Spam, abusive commentary, personal attacks, hate content or obvious rating manipulation may be removed without notice.
- Buying reviews, or trading them in developer groups, is coordinated manipulation.
- One review per player per listing is enforced by the data model: a second submission from the same account replaces the first rather than adding to it.
- Review text is capped at 2,000 characters and the rating is clamped to 1 to 5, so volume and extremes are the levers a manipulator has, and both are visible.
- Appsinity reserves the right to restructure, limit or temporarily suspend review features as abuse mitigation, without that counting as a service failure.
Manufactured demand
Traffic and conversion numbers come from recorded listing events, and each surface counts a defined thing: a page view, a play start for web titles, or reported seconds of play through the play-stats endpoint. There is currently no automated install-farming filter that discards suspicious events.
| Signal | How it is recorded today | What a farm changes |
|---|---|---|
| Listing views | An event row per view, with source and country | Inflates reach |
| Web opens | A play event for a web title | Inflates engagement |
| Play time | Reported seconds per session | Inflates retention story |
| Downloads | Counted from the download action | Inflates the headline number |
| Installs | Unique installs or unique web opens, labelled as such | Nothing distinguishes a bot |
Other spam surfaces
- Support tickets: duplicate tickets for the same issue delay the one that gets read; the assistant routes by topic rather than by urgency claimed.
- Team invites: inviting addresses you do not own to manufacture workspace breadth is a misuse of the member model.
- Webpages: Studio-hosted pages are subject to the same metadata rules as listings, see Studio webpages.
- Live streams: a schedule entry that is not an actual stream is treated as false information on the store, see Live streaming.
Frequently asked questions
Can I ask friends to review my launch?
You can ask for honest feedback. What you cannot do is pay, incentivise, organise or supply accounts for it. A handful of genuine friends is fine; a cluster of same-hour five-star reviews is the pattern that reads as coordination.
How do you detect fake installs?
Today the counters are recorded faithfully and anomalies are reviewed manually; there is no automated bot-filtration layer. See Downloads for what the metric actually counts.
Is cross-promoting my other games spam?
Not if the listing describes the game. A description that is mostly links to your catalogue is where it becomes a problem.
Someone is review-bombing me.
Report it with listing URLs and the pattern. Rating manipulation is explicitly removable content, see Reviews.
Developer publishing support: [email protected]
General support: [email protected]
Partnerships: [email protected]