Member Lifecycle Intelligence

Proactive member health tracking, risk detection, and retention intelligence for your Discord community

👑 ProPro tier feature

Key Features

Health Scoring

👑 Pro

Every member gets a real-time health score (0-100) based on 20+ behavioral features including engagement, social connections, consistency, and activity recency.

Lifecycle States

👑 Pro

9 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

👑 Pro

Automatic identification of at-risk members with risk families: inactivity drift, engagement cliff, social isolation, rapid decline, and ghost patterns.

Social Graph

👑 Pro

Maps member relationships through co-activity, replies, and reactions. Identifies isolated members and community connectors.

Cohort Analysis

👑 Pro

Track onboarding success by join-week cohorts. See which cohorts activate fastest and which have the highest churn rates.

Interventions

👑 Pro

Optional 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

1
Event Collection

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.

2
Feature Extraction

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.

3
Health & Score Computation

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.

4
State Assignment

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.

5
Risk & Intervention

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

MLI begins syncing your existing members immediately upon activation. Members receive an initial state based on their join date, then get progressively more accurate scores as activity data accumulates.

Lifecycle States

🆕
New

Just joined, no meaningful activity yet. Members start here and should activate within the first few days.

🌱
Activating

Showing early signs of engagement — first messages, first voice sessions. The critical onboarding window.

Active

Regularly active with consistent participation. Health score above engagement threshold.

Core

Your most valuable members. High engagement, strong social connections, consistent activity, and top-tier health scores.

🌊
Drifting

Activity declining from previous levels. An early warning signal — these members may still be recoverable with engagement.

⚠️
At Risk

Significant activity drop or extended absence. Risk patterns detected. Intervention recommended.

💔
Fragile

Critical state — very low engagement, multiple risk factors active. Requires immediate attention.

💤
Dormant

No activity for an extended period (configurable, default: 30 days). May return but unlikely without external triggers.

🔄
Revived

Previously inactive member who has returned and started engaging again. Re-entering the activation pipeline.

👻
Lost

Inactive beyond the lost threshold (configurable, default: 90 days). Statistically unlikely to return.

Note

State transitions are tracked with timestamps. You can see each member's previous state, when they transitioned, and how long they've been in their current state — useful for understanding churn patterns.

Risk Detection

MLI tracks 13 risk families, each detecting a distinct behavioral pattern that precedes member churn:

Engagement Decay

Gradual decline in activity frequency. The most common churn precursor — members slowly disengage before going silent.

Social Isolation

Low social reciprocity despite activity. The member participates but lacks meaningful connections with others.

Onboarding Failure

New member failed to activate within the expected window. They joined but never started engaging with the community.

Anchor Loss

A member's primary social connections or channels have become inactive, removing their reason to stay.

Schedule Mismatch

The member's active hours no longer overlap with peak community activity, reducing their engagement opportunities.

Moderator Friction

Recent moderation actions (warnings, timeouts) correlating with declining engagement.

Recognition Deficit

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

Each risk has a severity level (low, medium, high, critical). The /lifecycle at-risk command shows members sorted by risk severity, making it easy to prioritize outreach.

Getting Started

Via Discord Commands

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.

Via Dashboard

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

MLI requires the Pro tier. If your server is on the Free or Premium plan, the module will be available but locked. Upgrade to Pro to unlock lifecycle intelligence.

Frequently Asked Questions

What is Member Lifecycle Intelligence (MLI)?
MLI is a Pro-tier module that continuously tracks every member's lifecycle journey in your server — from the moment they join, through activation, engagement, and potential churn. It assigns health scores, detects risk patterns, and surfaces actionable insights so you can retain your most valuable community members.
How does the health score work?
Each member receives a health score from 0 to 100, calculated from 20+ behavioral features: message frequency, voice time, reaction patterns, social connections, consistency, and recency. The score updates on every compute cycle (configurable, default: 60 minutes) and tracks trends over rolling 7-day and 30-day windows.
What are lifecycle states?
MLI assigns each member one of 10 lifecycle states: New, Activating, Active, Core, Drifting, Fragile, At Risk, Dormant, Revived, and Lost. These reflect the member's current engagement trajectory and are powered by a 7-layer decision engine that considers activity patterns, social integration, and historical trends.
Does MLI read message content?
No. MLI only analyzes metadata — timestamps, counts, channel IDs, and behavioral patterns. It never stores or processes actual message content. All data is retained according to your configured policy and supports full GDPR/CCPA compliance.
How long until MLI has enough data to be useful?
MLI starts providing initial states immediately upon activation by analyzing join dates and recent activity. After 24-48 hours of data collection, health scores and risk assessments become increasingly accurate. Full confidence is typically reached after 7-14 days of continuous operation.
Can I configure thresholds and alerts?
Yes. Every key threshold is configurable: inactivity days, dormancy days, health alert thresholds, risk alert thresholds, compute intervals, and more. You can also set an alert channel to receive automated notifications when members enter critical states or when risk patterns are detected.