Health Tracking
💎 PremiumMLI computes a health score for every tracked member based on their activity, engagement depth, and consistency. Health scores power at-risk detection, interventions, and retention insights.
How Health Scores Work
Each member receives a health score from 0 to 100. The score is recalculated on every compute cycle (default: every 30 minutes). A higher score indicates a more active, engaged member. Scores factor in multiple signals weighted by recency.
Healthy (70–100)
Member is actively participating — sending messages, joining voice, reacting, and engaging with the community on a regular basis.
Moderate (40–69)
Member is present but showing reduced engagement compared to their baseline. Activity has declined but hasn't dropped to concerning levels.
At-Risk (0–39)
Member's activity has significantly declined. They may be on the path to becoming inactive or leaving the server entirely.
Health Score Factors
The health score is a weighted composite of several engagement signals. Each factor is normalized and weighted by recency — recent activity counts more than older activity.
Message Activity
Frequency and consistency of messages sent. Accounts for both volume and regularity — sporadic bursts score lower than steady daily engagement.
Voice Participation
Time spent in voice channels. Voice activity is a strong engagement indicator and carries significant weight in the health score.
Reaction & Interaction
Reactions added, threads participated in, and replies to other members. Passive engagement signals that indicate the member is still reading and interacting.
Recency Decay
All signals are weighted by recency. Activity from 7 days ago contributes less than activity from today. This ensures the score reflects current engagement, not historical peaks.
Viewing Health Data
/lifecycle overviewView aggregate health distribution across the entire server — how many members are healthy, moderate, or at-risk.
/lifecycle member <user>View a specific member's health score, trend direction, activity breakdown, and risk level.
/lifecycle new-members [days]View health scores for recently joined members to monitor onboarding effectiveness.
Compute Interval
/mli compute to trigger an immediate recalculation.Health Tracking Commands
| Command | Description |
|---|---|
/lifecycle overview | Server-wide health distribution |
/lifecycle member <user> | Individual member health data |
/lifecycle new-members [days] | New member onboarding health |
/mli compute | Manually trigger health recomputation |