HTTP event ingestion
POST /v1/events: batches of up to 500 events, idempotency by insert_id, a 200 response only after the write to ClickHouse.
A portfolio of all your apps, audience, retention and realtime — counted by people, not devices. One session_start event, and the app is already in your reports.
A fixed set of reports you need every morning. No report builder, no extra settings.
POST /v1/events: batches of up to 500 events, idempotency by insert_id, a 200 response only after the write to ClickHouse.
All apps on one screen: DAU, new users, sessions, D1/D7 and a 30-day trend. A morning digest in Telegram.
DAU/WAU/MAU, new users, sessions and session length. Breakdowns by version, OS, device, country and language, plus period comparison.
Classic D0–D30 and W0–W12 cohorts defined the same way as in AppMetrica. A heatmap and an average curve.
Users active right now, session starts per minute and a feed of the latest events that links to profiles.
No IP addresses stored — only country and city. Per-person data deletion on request, a deletion log and a PII guard for user_id.
Create an app, copy the write key and send your first batch. The format is the same for iOS, Android, web, bots and backends — no SDK needed.
curl -X POST https://i.analytics.example.com/v1/events \
-H "Content-Type: application/json" \
-H "X-App-Key: $AM_PK" \
-d '{
"sent_at": "2026-10-14T09:00:00Z",
"context": { "platform": "ios" },
"events": [{
"insert_id": "4b1c…",
"name": "session_start",
"timestamp": "2026-10-14T09:00:00Z",
"install_id": "5d8e…",
"session_id": "a0f1…"
}]
}'
→ 200 {"accepted":1,"rejected":[],"warnings":[]}Passkey-only sign-in.