Alerts & Notifications

Set up alert rules to get notified via Discord or email when critical metrics change. Alerts run automatically on a schedule and compare current values against a configurable baseline window.

Alert Types

Type	Trigger Condition	Use Case
Crash Spike	Error count exceeds X% of baseline	Detect crash loops after updates
Revenue Drop	Daily revenue falls below X% of baseline	Catch payment issues early
DAU Drop	Active users drop below X% of estimated average	Spot retention problems
Retention Drop	Day-7 retention of recent cohorts below X% of baseline	Onboarding regression
Log Error Spike	Log errors+warnings exceed X% of baseline	Remote logging anomalies

Creating an Alert Rule

1. Go to Configuration > Alerts
2. Click New Alert
3. Configure:
   • Name: Descriptive name (e.g. “Post-Update Crash Watch”)
   • Type: Select alert type
   • Threshold: Multiplier or percentage (e.g. 2.0 = trigger when 2x baseline)
   • Comparison Window: Baseline period (24h, 72h, 1 week, 1 month)
   • Email: Notification email address
   • Discord Webhook: Optional Discord webhook URL
4. Click Save

Threshold Logic

• Spike types (Crash Spike, Log Error Spike): Trigger when current value >= baseline * threshold. A threshold of 2.0 means “trigger at 200% of normal”.
• Drop types (Revenue, DAU, Retention): Trigger when current value < baseline * threshold. A threshold of 0.7 means “trigger when below 70% of normal”.

Discord Webhook Setup

1. In Discord: Server Settings > Integrations > Webhooks > New Webhook
2. Copy the webhook URL (starts with https://discord.com/api/webhooks/...)
3. Paste into the alert rule’s Discord Webhook field
4. Use Test Webhook to verify the connection

Alerts appear as rich embeds with color-coded types:

Type	Embed Color
Crash Spike	Red
Log Error Spike	Dark Red
Revenue Drop	Orange
DAU Drop	Blue
Retention Drop	Purple

Alert History

The History tab shows all triggered alerts with:

• Alert Name and type
• Change % from baseline
• Delivery Status (email sent, webhook sent, errors)
• Triggered At timestamp

Cooldown

Alerts have a 1-hour cooldown per rule to prevent notification spam. After triggering, the same rule won’t fire again for at least 60 minutes.

API Reference

# List alert rules for a game
curl "https://api.questdata.io/v1/alerts?game_id=GAME_ID" \
  -H "Authorization: Bearer YOUR_JWT"

# Create an alert rule
curl -X POST "https://api.questdata.io/v1/alerts" \
  -H "Authorization: Bearer YOUR_JWT" \
  -H "Content-Type: application/json" \
  -d '{
    "game_id": "GAME_ID",
    "name": "Log Error Watch",
    "type": "log_error_spike",
    "threshold": 2.0,
    "comparison_window_hours": 168,
    "email": "dev@example.com",
    "discord_webhook_url": "https://discord.com/api/webhooks/..."
  }'

# Get alert history
curl "https://api.questdata.io/v1/alerts/history?game_id=GAME_ID&limit=50" \
  -H "Authorization: Bearer YOUR_JWT"

# Test Discord webhook
curl -X POST "https://api.questdata.io/v1/alerts/RULE_ID/test-webhook" \
  -H "Authorization: Bearer YOUR_JWT"

Best Practices

1. Start with Crash Spike + Log Error Spike — these catch the most urgent issues
2. Use a 1-week comparison window (168h) for stable baselines
3. Set threshold to 2.0-3.0 for spikes — lower values cause false positives
4. Set threshold to 0.5-0.7 for drops — catch significant declines without noise
5. Connect Discord for instant visibility — email has delivery delays
6. Test your webhook before relying on it for production alerts