Analytics Alerts
Set up intelligent alerts to stay informed about important changes in your server's health, member activity, growth, and engagement metrics. Get notified when action is needed.
Alert Overview
Alerts are automated notifications triggered when specific conditions are met in your server analytics. Never miss important community changes.
Real-Time Notifications
Get notified instantly when alert conditions are triggered.
Customizable Thresholds
Set custom trigger points based on your community size.
Multiple Alert Types
Configure alerts for different metrics and events.
Channel Targeting
Send alerts to specific channels for organized notification.
Alert Scheduling
Control when alerts are active (e.g., business hours only).
Alert Cooldown
Prevent alert spam with configurable cooldown periods.
Alert Types
SYNTHET offers a variety of alerts to monitor different aspects of your community:
Member Drop Alert
Triggered when members leave at an abnormal rate. Helps catch churn issues early.
/alert add member-drop threshold=5 window=1hNotify if 5+ members leave within 1 hour.
Health Score Drop Alert
Alerts when your overall health score drops significantly. Indicates declining community health.
/alert add health-drop threshold=10 trigger=score-changeNotify if health score drops by 10 points in one check.
Low Engagement Alert
Triggered when engagement rate drops below a threshold. Detect participation decline early.
/alert add low-engagement threshold=35 period=7dAlert if weekly engagement rate drops below 35%.
Activity Spike Alert
Triggers when activity suddenly increases. Useful for trending events or viral moments.
/alert add activity-spike threshold=150% window=1hNotify when messages exceed 150% of average within 1 hour.
Bot Detection Alert
Alerts when patterns suggest bot spam or unwanted automation activity.
/alert add bot-detection sensitivity=highEnable AI-powered bot spam detection alerts.
Critical Health Alert
Triggers when health score enters critical range (below 25). Immediate action needed.
/alert add critical-health threshold=25 severity=criticalAlert when score drops below 25 with elevated severity.
Silent Channel Alert
Notifies when high-traffic channels go silent unexpectedly.
/alert add silent-channel threshold=0 window=6hAlert if typically active channel has no messages in 6 hours.
Milestone Alert
Celebrate important growth milestones like member count or activity records.
/alert add milestone trigger=members-reached value=1000Celebrate reaching 1000 members or custom milestones.
Creating Alerts
Set up custom alerts with specific thresholds and conditions:
Basic Alert Command
/alert add <alert-type> [options]alert-type: Type of alert to create (member-drop, health-drop, etc.)
threshold: Value that triggers the alert
window/period: Time window for evaluating the alert
channel: Which channel to send notifications to
cooldown: Minimum time between alerts (prevents spam)
Example: Member Drop Alert
/alert add member-drop threshold=10 window=2h channel=#alerts cooldown=30mNotifies in #alerts if 10+ members leave within 2 hours, with alerts no more than every 30 minutes.
Example: Low Engagement Alert
/alert add low-engagement threshold=40 period=7d channel=#admin-alerts cooldown=24hNotifies in #admin-alerts if weekly engagement drops below 40%, checked once per day.
Example: Health Score Alert
/alert add health-drop threshold=15 severity=high channel=#leadershipNotifies leadership if health score drops by 15 points (marked as high severity).
Managing Alerts
Control and configure your alerts:
List All Alerts
/alert listView all configured alerts, their status, and recent triggers.
Edit Alert
/alert edit <alert-id> threshold=<new-value>Modify existing alert settings like thresholds or target channels.
Disable/Enable Alert
/alert toggle <alert-id>Temporarily disable or re-enable alerts without deleting them.
Delete Alert
/alert delete <alert-id>Remove an alert permanently.
View Alert History
/alert history <alert-id>See when alerts were triggered, what values triggered them, and notification history.
Set Alert Preferences
/alert preferences [option]Configure global alert settings like quiet hours, default channels, or notification style.
Alert Routing
Control where alerts are sent based on severity and type:
Channel-Based Routing
Send different alert types to different channels:
/alert add low-engagement channel=#engagement-alerts/alert add critical-health channel=#leadershipRole-Based Notifications
Mention specific roles when alerts trigger:
/alert add member-drop notify-role=@ModeratorsDM Notifications
Send critical alerts as direct messages to server admins:
/alert add critical-health notify-dm=adminsSummary Reports
Get daily or weekly summaries instead of real-time alerts:
/alert add summary report=daily channel=#weekly-summaryRecommended Alert Setup
Here's a recommended set of alerts for most servers:
For Small Servers (50-500 members)
- Member drop alert: 3+ members in 4 hours
- Health score drop: 15+ points
- Low engagement: Below 30% weekly
- Milestone: Every 100 members
For Medium Servers (500-5000 members)
- Member drop alert: 10+ members in 2 hours
- Health score drop: 20+ points
- Low engagement: Below 40% weekly
- Activity spike alert: 200% in 1 hour
- Critical health: Below 25
For Large Servers (5000+ members)
- Member drop alert: 50+ members in 1 hour
- Health score drop: 25+ points
- Low engagement: Below 45% weekly
- Activity spike/drop: 150%+ or 50% in 30 mins
- Bot detection: High sensitivity
- Silent channel: 12+ hours for top 5 channels
Alert Best Practices
- • Set thresholds based on your server's baseline activity, not arbitrary numbers
- • Use cooldown periods to prevent alert fatigue from spammy triggers
- • Test alerts with lower thresholds first, then adjust to find the sweet spot
- • Route critical alerts to leadership channels and moderate alerts to general channels
- • Review alert history regularly to understand what triggers false positives
- • Disable alerts temporarily during special events that will skew metrics
- • Use alerts to catch problems early, not as a reaction system
- • Adjust thresholds seasonally if your server has cyclical activity patterns
- • Act on alert notifications quickly - delayed response reduces effectiveness
- • Document why each alert exists so future moderators understand their purpose