Fire an event
Send an authenticated event name, device token, properties, and template variables.
Event-triggered automation
Respond when the game creates the moment.
Turn authenticated backend events into durable, delayed, and observable player notifications without blocking gameplay services.
Illustrative trigger activityTrigger active
Overview
Audience
Messages
Delivery
Settings
Live workspace
guild.member_request
Delay
15 minMatches today
842Retries
BoundedCurrent flow
Event acceptedIdempotency verified
Template resolvedguild-invite-v2
Dispatch queuedPlayer policy applied
Example data for product illustration. Not customer performance.
Ingress should be fast. Delivery should be durable.
PushHub acknowledges a game event after durable storage, then performs provider work through the same recoverable delivery engine as campaigns.
How it works
A clean boundary between your game and push providers.
Match triggers
Resolve active rules and store token-free durable jobs after validation.
Wait safely
Apply the configured delay without holding open a game-service request.
Deliver and recover
Lease work, record attempts, retry transient failure, and dead-letter permanent issues.
Capability depth without the fine-print surprise.
Current and planned capabilities are labeled directly, so teams can evaluate PushHub without confusing vision for availability.
Published event endpoint
Fire app-scoped events through a documented API with scoped-key authorization.
Idempotent ingestion
Hash an optional idempotency key before storage to prevent duplicate trigger work.
Durable delayed jobs
Store dispatch intent in PostgreSQL and use Redis only as a wake-up path.
Visual multi-step journeys
A planned journey builder will connect several conditions, waits, messages, and exits.
In the game
Always-on moments without hidden machinery.
Reward ready
Notify a player when crafting, energy, or a timed reward becomes available.
Social pull
Bring a player back when a guild or friend action needs attention.
Purchase recovery
Follow a trusted server-side payment event with a helpful next step.
Automation needs an audit trail, not magic.
Every trigger retains source identity, company and app ownership, normalized failure state, and secret-free attempt history.
Token-free jobs
Raw provider tokens are resolved before durable event payload storage.
Bounded retries
Transient failures stop after a defined retry budget.
Operator visibility
Keme administrators can see leases, failures, attempts, and worker freshness.
Questions teams ask before they commit.
No. The endpoint durably records matching work and provider submission happens asynchronously.
Build the next player return.
Start with one app, one meaningful moment, and one audience worth reaching.