Member Lifecycle Intelligence
Proactive member health tracking, risk detection, and retention intelligence for your Discord community
Key Features
Health Scoring
👑 ProEvery member gets a real-time health score (0-100) based on 20+ behavioral features including engagement, social connections, consistency, and activity recency.
Lifecycle States
👑 Pro9 distinct lifecycle states from NEW to LOST, assigned by a 7-layer decision engine that adapts to your community's unique activity patterns.
Risk Detection
👑 ProAutomatic identification of at-risk members with risk families: inactivity drift, engagement cliff, social isolation, rapid decline, and ghost patterns.
Social Graph
👑 ProMaps member relationships through co-activity, replies, and reactions. Identifies isolated members and community connectors.
Cohort Analysis
👑 ProTrack onboarding success by join-week cohorts. See which cohorts activate fastest and which have the highest churn rates.
Interventions
👑 ProOptional automated alerts when members enter critical states. Configure an alert channel to receive real-time notifications with contextual explanations.
How MLI Works
The MLI Pipeline
MLI passively observes events from all other modules — messages, voice sessions, reactions, role changes, moderation actions, and more. Each event is normalized into a typed MLI event with category, action, and value.
20 behavioral features are computed per member across multiple time windows (1d, 7d, 14d, 30d): message count, voice minutes, reaction ratio, channel diversity, peak-hour activity, burst patterns, consistency, social reciprocity, and more.
Features are combined into composite scores: engagement score, social score, importance score, and overall health score. Each dimension is weighted and normalized to the 0-100 range with trend tracking.
A 7-layer decision engine evaluates each member: forced overrides, inactivity detection, re-engagement signals, core qualification, engagement gates, activation progress, and drift detection. The result is one of 10 lifecycle states.
Risk patterns are evaluated and tracked as active risks. When interventions are enabled, state transitions to critical states trigger alerts in your configured channel with context and recommendations.
Tip
Lifecycle States
Just joined, no meaningful activity yet. Members start here and should activate within the first few days.
Showing early signs of engagement — first messages, first voice sessions. The critical onboarding window.
Regularly active with consistent participation. Health score above engagement threshold.
Your most valuable members. High engagement, strong social connections, consistent activity, and top-tier health scores.
Activity declining from previous levels. An early warning signal — these members may still be recoverable with engagement.
Significant activity drop or extended absence. Risk patterns detected. Intervention recommended.
Critical state — very low engagement, multiple risk factors active. Requires immediate attention.
No activity for an extended period (configurable, default: 30 days). May return but unlikely without external triggers.
Previously inactive member who has returned and started engaging again. Re-entering the activation pipeline.
Inactive beyond the lost threshold (configurable, default: 90 days). Statistically unlikely to return.
Note
Risk Detection
MLI tracks 13 risk families, each detecting a distinct behavioral pattern that precedes member churn:
Gradual decline in activity frequency. The most common churn precursor — members slowly disengage before going silent.
Low social reciprocity despite activity. The member participates but lacks meaningful connections with others.
New member failed to activate within the expected window. They joined but never started engaging with the community.
A member's primary social connections or channels have become inactive, removing their reason to stay.
The member's active hours no longer overlap with peak community activity, reducing their engagement opportunities.
Recent moderation actions (warnings, timeouts) correlating with declining engagement.
Active contributor receiving little recognition or positive feedback relative to their contribution level.
Plus: Event Detachment, Role Mismatch, Content Mismatch, Support Frustration, Competitive Displacement, and Relapse.
Tip
/lifecycle at-risk command shows members sorted by risk severity, making it easy to prioritize outreach.Getting Started
1. Enable the module — Go to your dashboard and enable the MLI module from the Modules page, or use the bot's module management command.
2. Configure settings — Use the dashboard to set your alert channel, thresholds, and enable interventions.
3. Check status — /mli status to verify MLI is running and collecting data.
4. View your server — /lifecycle overview to see the guild-wide health summary.
5. Check at-risk members — /lifecycle at-risk to see who needs attention.
1. Enable MLI — Navigate to your server's dashboard, click "Modules" in the sidebar, and toggle MLI on.
2. Open the MLI page — In the dashboard sidebar, click "Member Lifecycle" to access the MLI dashboard.
3. Configure settings — Set your alert channel, thresholds, and enable interventions from the dashboard config panel.
4. Browse members — Use the member table to sort by health score, filter by state, and drill into individual member profiles.
5. Monitor trends — The overview cards show guild-wide health, state distribution, at-risk count, and recent state transitions.
Warning