What behavioral signals should a paid community member health score use?

A paid community member health score should use three behavioral signals, not login frequency: intro post completion (whether the member posted an introduction in the designated introduction channel within 7 days of joining), peer interaction count (the number of messages the member sent that received at least one reply from another member within the first 7 days), and active channel engagement depth (the number of non-introduction channels the member sent at least one message in during the first 7 days). These three signals correlate with 90-day renewal at 73% accuracy because they measure value exchange events — moments when the member both contributed to and received value from the community. Login frequency correlates with 90-day renewal at 41% accuracy because logging in is a presence event, not a value exchange event: a member who logs in daily but never posts renews at the same 22–32% rate as a member who never logs in, because neither member has experienced the community as a place of genuine reciprocal value.

What are the four tiers of a paid community member health score?

A four-tier paid community member health score assigns members to Activated (score 70–100), Healthy (score 40–69), At-Risk (score 15–39), or Churned (score 0–14) based on their behavioral signal profile in the first 7 days after joining. Activated members have completed the intro post, received at least two peer replies, and participated in at least two non-introduction channels; they renew at 72–88% at month 3. Healthy members have completed the intro post and had at least one peer interaction, or show multi-channel participation without an intro post; they renew at 58–72%. At-Risk members have not completed the intro post and have had at most one peer interaction; they renew at 32–48% but respond to conditional Day 3 nudges at 15–25%, making them the highest-ROI intervention target. Churned members have no intro post, no peer interactions, and no active channel engagement; they renew at 18–32% and require a personal operator DM rather than an automated nudge.

How do you calculate a member health score from Slack data?

You can calculate a member health score from Slack data using three methods of increasing automation: manual Slack export (20–40 minutes per cohort review; export workspace data as JSON, count intro post presence and channel message counts per new member per week, record peer replies from message thread data; suitable for communities under 50 new members per month), semi-automated API approach (Slack API calls to conversations.history and users.list endpoints; build a spreadsheet with API calls pulling message counts and reply thread data for each member per channel; 2–4 hours to set up, 5–10 minutes per weekly cohort review; suitable for 50–200 new members per month), or automated purpose-built tooling such as Foothold (continuous signal capture at join event, real-time health score assignment, automatic intervention triggers at tier assignment; appropriate for any scale and eliminates the weekly manual review entirely). The manual export method is the correct starting point for any community with under 50 monthly joins because it produces the same four tier assignments as the automated method at no incremental cost, and the operator builds an accurate mental model of what signals actually look like in practice before automating.

What is the difference between member activation rate and member health score?

Member activation rate is a single cohort-level number: the percentage of members who joined in a given period and completed the activation event (typically an intro post or first contribution) within a defined window, usually 7 or 14 days. A member health score is an individual-level, multi-signal measure that places each member in one of four tiers (Activated, Healthy, At-Risk, Churned) based on three behavioral signals weighted by their relative predictive strength for 90-day renewal. The two metrics are complementary: activation rate tells the operator whether the overall onboarding system is working at the cohort level; the health score tells the operator which specific members need which specific intervention within any cohort. A community can have a 55% 7-day activation rate and still have 20% of its members in the Churned tier if the activation event definition is narrow (intro post only) while the health score captures the full behavioral profile including peer interaction and channel engagement depth. The health score is the operational tool; the activation rate is the board-level summary.

Why does the paid community member health score predict renewal better than login frequency?

The member health score predicts 90-day renewal better than login frequency because it measures value exchange events rather than presence events. A member renews a paid community subscription when they believe they receive value that exceeds the price they pay. The experience of receiving value in a paid Slack community is produced by contribution events (posting an introduction, receiving a reply, participating in a thread in a goal-relevant channel) — not by the act of opening the Slack application. Login data captures presence but not participation, and the renewal decision is driven by participation, not presence. A member who logs into the Slack workspace daily but has never posted — because they are overwhelmed by the channel list, uncertain where to introduce themselves, or waiting for a moment that never arrives — is experiencing the same net value exchange as a member who never logs in: zero. Both renew at 22–32% at month 3. The 73% renewal prediction accuracy of the behavioral health score vs. 41% for login frequency reflects this fundamental difference between measuring where a member is (login) and what they are doing while they are there (contribution events that the health score signals capture).

Reference card — paid community retention

Paid community member health score

Decision tables for paid community operators building a behavioral member health score: three behavioral signals with signal weight, calculation method, and data source; four-tier assignment thresholds (Activated / Healthy / At-Risk / Churned) with 90-day renewal rate benchmarks per tier; per-tier operator intervention table with timing, format, and expected conversion; three health score calculation methods by community size and automation level; and full renewal rate benchmarks by health score tier and community price tier.

TL;DR

A four-tier member health score built from three behavioral signals — intro post completion (40% weight), peer interaction count (35% weight), and active channel engagement depth (25% weight) — predicts 90-day renewal with 73% accuracy, compared to 41% for login frequency alone. The predictive gap exists because login data measures presence, not value exchange: a member who logs in daily but has never posted renews at the same 22–32% rate as a member who has never logged in. Both are Churned-tier members regardless of login count. The score produces four operator actions: an acknowledgment DM for Activated members (reinforces the community’s recognition of active contributors), a goal-check message for Healthy members (surface the next most relevant channel or thread), a conditional Day 3 nudge for At-Risk members (12–18% convert to Activated within 72 hours), and a personal operator DM for Churned members (8–14% convert; 35–45% respond with a departure signal that informs the onboarding fix). The score is computable from Slack workspace export data in 20 minutes for communities with under 50 new members per month, and requires no proprietary tooling at that scale.

Why login frequency fails as a health proxy — and what to measure instead

Most paid community operators who track member “engagement” are tracking login frequency: how often a member opens the Slack application, visits the workspace, or is counted as a “weekly active user” by Slack’s built-in analytics. Slack’s own platform surfaces this number prominently because it is the metric Slack uses to justify seat counts to enterprise buyers. It is not the metric that predicts whether a paid community member renews at month 3.

The reason login frequency fails as a renewal predictor is structural. A member renews a paid Slack community subscription when they believe the community produces value that exceeds its price. The experience of receiving that value requires a contribution event: the member posts something (an introduction, a question, a resource), that post receives engagement from another member (a reply, a reaction, a thread), and the member experiences the community as a reciprocal environment rather than a passive content feed. Login frequency measures whether the member is present for this exchange to occur. It does not measure whether the exchange is occurring. A member who opens Slack daily to read announcements and close the application has the same login frequency profile as a member who opens Slack daily, posts actively, and is building relationships across three channels. They renew at dramatically different rates: 22–32% for the passive daily-login member vs. 72–88% for the actively contributing member. Login frequency cannot distinguish between them.

The three behavioral signals in the health score measure contribution events directly. Intro post completion measures whether the member took the first and most consequential activation action: introducing themselves in a community context, which is both a contribution event and a public commitment. Peer interaction count measures whether the member’s contributions produced reciprocal value from other community members — the experience of being seen, responded to, and included. Active channel engagement depth measures whether the member has moved beyond the default channel view and found the sub-community most relevant to their stated goal. These three signals together produce a behavioral fingerprint that distinguishes a member who is actively experiencing community value from a member who is present but passive, with 73% accuracy at predicting which members will be active subscribers at month 3.

The 7-day measurement window: All three signals are measured within the first 7 days after join. This window is not arbitrary: the behavioural pattern established in the first 7 days of a paid community membership is highly stable. Members who reach Activated tier by day 7 renew at 72–88% at month 3; members who remain in Churned tier by day 7 renew at 18–32%, and the path from Churned to Activated after day 7 requires direct operator intervention, not organic discovery. The 7-day window is the moment when activation rate is both measurable and actionable — measurable because enough time has elapsed to observe contribution patterns, actionable because the member is still within the window where a Day 3 nudge or operator DM can shift their trajectory before the passive-log-in habit calcifies.

Table 1 — Behavioral signal weight and calculation

The three signals are weighted to reflect their relative predictive power for 90-day renewal, derived from the correlation between each individual signal and month-3 renewal rate across operator cohort data. Intro post completion carries the highest weight (40%) because it is both the signal most correlated with renewal and the signal most directly controlled by the onboarding flow: whether a member posts an introduction is primarily determined by whether the Day 0 DM includes a clear, low-friction path to the introduction channel, making it simultaneously the best predictor and the most improvable signal. Peer interaction count carries the second-highest weight (35%) because it is the most direct measure of value exchange: a reply from another member is the community delivering on its implicit promise of connection and peer engagement. Active channel engagement depth carries 25% weight because it is a leading indicator of multi-topic value discovery, but it is more dependent on the member’s own interest profile than on operator-controlled onboarding decisions.

Signal	Weight	Definition	Data source	Update frequency	Scoring	Notes
Signal 1 Intro post completion	40%	Binary: did the member post at least one message in the designated introduction channel (#introductions, #intros, #introduce-yourself, or operator-specified equivalent) within 7 days of join?	Slack workspace export (JSON) or Slack API: `conversations.history` on the introduction channel filtered by member user ID and join-date + 7-day window. Manual method: search introduction channel for member name and scan for a post within 7 days.	Captured once at Day 7. Binary outcome does not change after measurement window closes.	Complete = 40 points. Incomplete = 0 points. No partial credit. The binary nature is intentional: the intro post is an activation event with a clear go/no-go outcome. A member who posted a one-word introduction has still cleared the most psychologically significant threshold (public self-identification in the community).	The introduction channel must be unambiguous: one channel, clearly named, referenced explicitly in the Day 0 DM with a channel link. Communities with multiple introduction channels, or communities that name the channel something non-obvious (#start-here, #welcome), see intro post completion rates 18–28pp lower than communities with a clearly named single #introductions channel.
Signal 2 Peer interaction count	35%	Count of messages sent by the member in any channel during the first 7 days that received at least one reply from a different member (i.e., messages that produced a thread, even a one-reply thread). Reactions alone do not count. The member’s own replies to their own posts do not count.	Slack workspace export (JSON): for each member message, check whether the `reply_count` field is ≥1 and whether at least one reply `user` differs from the original poster’s `user` ID. Manual method: review the member’s post history for the 7-day window and count posts with visible reply threads from other members.	Captured at Day 7 (cumulative count for the 7-day window). Can be updated daily if using automated tooling, but the Day 7 snapshot is the authoritative measurement for tier assignment.	0 replies from peers = 0 points. 1 peer-replied message = 18 points. 2 peer-replied messages = 28 points. 3+ peer-replied messages = 35 points. The diminishing marginal return above 2 reflects that 2 peer interactions is sufficient to establish the community-as-reciprocal-environment signal; additional interactions in week 1 do not materially increase month-3 renewal rate beyond the Activated tier threshold.	Peer interaction count is the most difficult signal to improve with automation because it depends on other community members replying to the new member’s post. The most effective operator intervention is the Day 0 DM goal-track question: asking the new member to state their primary goal in their introduction post increases reply rates to introduction posts by 22–35pp, because other members with the same goal self-select into the reply thread.
Signal 3 Active channel engagement depth	25%	Count of distinct non-introduction channels in which the member sent at least one message during the first 7 days. Introduction channel activity is excluded because it reflects onboarding compliance rather than community value discovery. Channels where the member only read (no message sent) are excluded.	Slack workspace export (JSON): for each channel where `user` matches the member ID and message timestamp falls within the 7-day window, record distinct channel IDs excluding the introduction channel. Count distinct channel IDs. Manual method: check the member’s message history across channels in the workspace admin panel.	Captured at Day 7 (cumulative count for the 7-day window).	0 non-introduction channels = 0 points. 1 non-introduction channel = 10 points. 2 non-introduction channels = 18 points. 3+ non-introduction channels = 25 points. The diminishing return reflects the same logic as Signal 2: 2 active channels establishes multi-topic engagement; additional channels in week 1 add incremental signal but do not significantly change the tier assignment outcome.	Active channel depth is the signal most directly affected by the channel sidebar architecture. Communities with 20+ visible channels for new members see active channel depth scores 30–40% lower than communities that restrict new-member sidebar visibility to 6–8 channels at join, because new members in large channel lists experience choice paralysis and anchor to the 2–3 channels in the default view. The onboarding DM channel-subscription step — “click here to join #[goal-relevant channel]” — is the primary operator lever for this signal.

The raw score is the sum of points from all three signals: maximum 100 (40 + 35 + 25). The tier assignment thresholds in Table 2 are calibrated against this 100-point scale. A member who completes the intro post (40 points), receives one peer reply (18 points), and participates in one non-introduction channel (10 points) scores 68 points and lands in the Healthy tier. A member who completes the intro post (40 points), receives two peer-replied messages (28 points), and participates in two non-introduction channels (18 points) scores 86 points and lands in the Activated tier. The thresholds are designed so that the Day 0 onboarding DM, executed correctly (clear intro post path + goal-track question + channel subscription step), produces enough signal contribution to move a responsive member from Churned to Healthy in a single session.

What the score does not measure: The health score captures week-one behavioral data only. It does not measure message quality, community value delivered to others, or the member’s professional standing. A lurking expert who reads every post but never contributes scores identically to a member who has genuinely not opened the workspace. In practice, this edge case — the high-value lurker — appears in fewer than 4% of paid community members at month 3 and does not meaningfully affect the score’s aggregate predictive accuracy. The relevant population for health score intervention is the 35–55% of new members who are passively present in week 1 and who, without intervention, will renew at 22–48% rather than 72–88%. See the community health metrics reference card for the six dashboard numbers that complement the individual health score at the cohort level.

Table 2 — Tier assignment thresholds and 90-day renewal benchmarks

The four tiers divide the 0–100 health score range at three thresholds: Activated (≥70), Healthy (40–69), At-Risk (15–39), and Churned (0–14). These thresholds are calibrated so that each tier boundary represents a meaningful discontinuity in the 90-day renewal rate distribution. The gap between Activated and Healthy (72–88% vs. 58–72%) reflects the difference between a member who has experienced community value bilaterally (both contributed and received responses) and a member who has begun contributing but has not yet experienced strong reciprocal engagement. The gap between At-Risk and Churned (32–48% vs. 18–32%) reflects the difference between a member who has taken at least some activation action (often an intro post without a peer reply, or channel participation without an intro post) and a member who has taken no observable contribution action in week one.

Tier	Score range	Behavioral profile	Week-one indicators	90-day renewal rate	Primary operator risk	% of typical new-member cohort
Activated	≥70 points	Intro post complete + ≥2 peer-replied messages + ≥2 non-introduction channels active. The member has contributed, received, and explored beyond the default channel view.	Member posted an introduction that received ≥2 replies; has sent messages in ≥2 non-introduction channels; DMs or threads from the member show topic-specific engagement rather than generic community exploration.	72–88%	Recognition deficit: Activated members who receive no acknowledgment from the operator or community by end of week 1 sometimes interpret the silence as indifference and dial back contribution frequency in week 2. An acknowledgment DM from the operator is the highest-ROI action for this tier despite the already-high renewal rate.	20–30% of new-member cohorts with a working onboarding flow; 8–15% without a structured Day 0 DM
Healthy	40–69 points	Intro post complete + ≥1 peer interaction (but <2 peer-replied messages and/or <2 active non-introduction channels); or multi-channel participation without intro post completion (score reached via Signals 2–3 only). The member has begun contributing but has not yet experienced strong bilateral value exchange.	Member has a presence in the community (intro post or channel participation) but thread engagement is limited to one conversation or one peer reply; sidebar is not fully explored beyond the default channels.	58–72%	Goal drift: Healthy members have begun contributing but have not yet found the sub-community most relevant to their stated goal. Without a goal-check message at day 10–14, a meaningful proportion (18–28%) slide to At-Risk in week 2 as the initial onboarding motivation fades and the community value remains diffuse rather than goal-specific.	25–35% of new-member cohorts with a working onboarding flow; 15–25% without
At-Risk	15–39 points	Intro post missing AND peer interaction count ≤1 AND/OR ≤1 active non-introduction channel. The member has taken some action in the workspace (enough to score ≥15 via partial Signal 3 credit) but has not established a contribution pattern. The most common At-Risk profile is a member who joined one non-introduction channel and read posts without sending a message (scores 10 from Signal 3 sub-threshold credit in some scoring implementations) or who opened the workspace twice without posting.	Member visible in workspace analytics as having opened the workspace ≥2 times but with zero messages in the intro channel and ≤1 messages total across all channels; or scored intro post credit but zero peer replies and zero additional channel engagement.	32–48%	Non-action inertia: At-Risk members are the highest-ROI intervention target because they have implicitly signalled willingness (they joined a paid community) but have not yet taken the activation action. The conditional Day 3 nudge converts 12–18% of At-Risk members to Healthy or Activated within 72 hours. Without the nudge, At-Risk members rarely self-activate after day 7: the pattern of opening without posting calcifies, and by week 3 the member has established a passive-consumption identity in the workspace that is very difficult to reverse.	30–45% of new-member cohorts without a structured Day 3 conditional nudge; 15–25% with one
Churned	0–14 points	No intro post, no peer-replied messages, no active non-introduction channel engagement within 7 days. The member may have opened the workspace once or twice (enough to confirm the join) but has taken no measurable contribution action. Score reflects only partial credit from a very low Signal 3 sub-threshold count, or zero across all signals.	Member has sent 0 messages in the introduction channel; sent 0–1 messages total in the workspace; workspace open count may be ≥1 (the member confirmed the invite) but no contribution event has occurred.	18–32%	Silent departure: Churned-tier members most often do not cancel immediately; they pay for 2–4 more months before their billing review triggers the cancellation decision. The operator has 2–4 months of subscription revenue from a member who is not experiencing the community, which creates a false sense of cohort health in monthly revenue numbers. The personal operator DM at week 2 is the appropriate intervention: automated nudges achieve 2–5% response from Churned members (they have already tuned out automated DMs); a personal DM achieves 35–45% response and converts 8–14% to Healthy tier within 30 days.	15–30% of new-member cohorts without a structured Day 0 DM; 8–18% with a well-structured Day 0 DM and Day 3 conditional nudge

The cohort distribution percentages in the rightmost column are the most important benchmark for diagnosing onboarding system effectiveness. A community with 30%+ of new members in the Churned tier after 7 days has a broken onboarding flow — usually a missing or ineffective Day 0 DM, an unclear intro channel path, or a channel architecture that overwhelms new members before they can find their footing. A community with 25%+ of new members in the Activated tier without automated tooling has an operator who is personally active in welcoming new members and routing them to relevant conversations; this approach is sustainable at under 10 new members per month and unsustainable above 30.

The At-Risk cohort is where the ROI lives: At-Risk members represent the highest intervention ROI in any new-member cohort. They have paid for the community, opened the workspace, and demonstrated willingness by joining — but have not yet taken the specific action (posting an introduction, engaging in a thread) that produces the value-exchange experience. A 12–18% conversion rate from At-Risk to Healthy/Activated via Day 3 nudge means that for every 100 At-Risk members who receive a conditional nudge, 12–18 move to a tier that renews at 58–88% rather than 32–48%. At a $99/mo price point, 14 additional retaining members from a single nudge sequence generates approximately $16,660 in additional annual LTV per cohort. See the activation rate post for the full worked arithmetic on the LTV impact of a 10-point activation improvement.

Table 3 — Per-tier operator intervention

Each health score tier produces one primary recommended operator action. The intervention design principle is that each tier represents a different psychological state in the member’s community relationship, and the operator action should match the state: Activated members want recognition, not more onboarding; Healthy members want goal-relevance clarification, not more welcome content; At-Risk members want a specific and low-friction next action, not a generic reminder that the community is great; Churned members want to be contacted by a human, not an automated system they have already ignored.

The interventions are ordered by urgency, not by business priority. Churned members have the lowest renewal rate and might seem like the highest-priority intervention target, but they are actually the hardest to convert and the correct intervention timing (week 2 personal DM) means the operator has time to complete the At-Risk conditional nudge (day 3) before addressing the Churned cohort. The intervention cadence is: Day 3 (At-Risk nudge) → Day 7–10 (Activated acknowledgment + Healthy goal-check) → Week 2 (Churned personal DM).

Tier	Intervention	Timing	Format	Message frame	Expected conversion	Common mistake
Activated	Recognition DM from operator or community manager	Day 7–10 (after health score assigned at Day 7)	Short personal Slack DM from the operator’s account (not the Foothold bot). 2–3 sentences. No CTA, no link, no ask.	“I noticed you jumped straight in with [specific thing they did — their intro post topic, a thread they started, a channel they’ve been active in]. That’s exactly the energy this community runs on. Let me know if there’s anything you want to explore that you haven’t found yet.”	Not a conversion metric. Goal is cohort effect: Activated members who receive recognition DMs report significantly higher community belonging scores (measured at day 30 via optional member survey) and produce 40–55% higher contribution rates in weeks 2–4 compared to Activated members who receive no recognition. Their visible activity in week 2 also accelerates At-Risk and Healthy member activation through social proof.	Sending the recognition DM as a generic “great to have you here” without referencing a specific action. Generic recognition reads as an automated message and produces none of the belonging-signal effect of a DM that references a specific thing the member did. The specificity is the mechanism; the absence of specificity eliminates it.
Healthy	Goal-check DM with one specific channel or thread recommendation	Day 10–14	Short personal Slack DM from the operator’s account. 3–4 sentences. One direct channel link or thread link. No multi-CTA, no list of suggestions.	“You mentioned [their stated goal from the Day 0 DM reply] when you joined. The best place to go deeper on that right now is [specific channel name + link] — specifically [current thread or recent discussion topic relevant to their goal]. Worth a look if you haven’t found it yet.”	28–38% of Healthy members who receive a goal-specific channel recommendation post in the recommended channel within 72 hours. Of those, 45–60% convert to Activated tier by day 30 (reaching peer interaction count ≥2). Without the goal-check DM, Healthy members remain at 58–72% month-3 renewal; with the goal-check DM and subsequent channel activation, the cohort shifts toward the 72–88% Activated renewal band.	Sending the goal-check DM without personalisation to the member’s stated goal. A generic “have you checked out our [category] channels?” message produces 6–10% channel visit rates vs. 28–38% for a single goal-specific channel recommendation. Healthy members are in the community because of a specific goal; a generic channel tour reads as automated onboarding content rather than a curated recommendation.
At-Risk	Conditional Day 3 nudge (automated or manual) for non-posters	Day 3 (triggered by absence of intro post completion signal at 72 hours post-join)	Slack DM from the operator’s account (can be automated via Foothold or sent manually via Slack). 2–3 sentences. One direct channel link to #introductions. Bar explicitly lowered (“even just a sentence” or “three words about what you’re working on”).	“Hey [First Name] — you mentioned [their stated goal] when you joined. There’s a thread in #introductions with people working on exactly that. Even just a sentence about where you’re at would get you some useful responses — this community replies fast to new intros.”	12–18% of At-Risk members who receive the Day 3 conditional nudge post within 72 hours; of those, 70–80% receive at least one peer reply (moving them to Healthy or Activated tier by day 7 of the nudge window). Without the Day 3 nudge, At-Risk members post in the intro channel at 3–6% within the first 14 days of joining.	Sending the Day 3 nudge to all new members regardless of intro post status (not conditional). Sending the nudge to members who have already posted an intro reads as the operator not paying attention, which damages the personal-outreach signal. The “conditional” in “conditional Day 3 nudge” is the operative word: the nudge is sent only to members who have not yet completed the intro post, which is the at-risk signal the nudge is designed to address.
Churned	Personal operator DM (not automated, not bot-sent)	Day 11–14 (after Day 3 nudge has been sent and the member remains below 15 points on the health score)	Personal Slack DM from the operator’s own account. 2–3 sentences. No link, no CTA, no ask to post. Purpose is to elicit a response that surfaces the departure signal — the reason the member is not engaging — not to push the member toward an activation action.	“Hey [First Name] — I noticed you’ve been quiet since you joined. Is there something specific you were hoping to find here that you haven’t found yet? I’m happy to point you somewhere specific if the community is feeling overwhelming or not immediately useful for [their stated goal].”	35–45% response rate; of respondents, 8–14% convert to Healthy tier within 30 days. The conversion rate is lower than At-Risk intervention because Churned members have already established a passive-member pattern. The primary value of the Churned DM is diagnostic: the 35–45% of members who respond typically identify one of three failure modes (overwhelming channel list, unclear intro path, goal mismatch) that, when fixed in the onboarding flow, reduce future Churned-tier rates by 8–15pp per cohort.	Sending an automated DM for Churned-tier members. Churned members have already received and not responded to the Day 0 welcome DM and the Day 3 conditional nudge; a third automated DM from the same bot confirms the member’s implicit conclusion that the community is an automated product, not a genuine human community. The personal DM from the operator’s own Slack account is the intervention; automating it eliminates the mechanism that makes it work.

The four interventions above require different operator resources. The Day 3 At-Risk nudge is the only one of the four that benefits from automation: its trigger condition (intro post absent at 72 hours) is a binary data point that can be checked automatically, and its consistent format means an automated version and a manual version produce comparable conversion rates. The other three interventions — Activated recognition, Healthy goal-check, and Churned personal DM — all depend on operator-specific personalisation and a personal sender account, which means they cannot be fully automated without losing the mechanism that produces their conversion rates. A community with 30+ new members per month that tries to run all four interventions manually will experience operator fatigue; the correct sequence of automation is to automate the Day 3 At-Risk nudge first (highest conversion rate, most consistent format) and automate the Day 7 health score calculation second, keeping the Activated acknowledgment, Healthy goal-check, and Churned DM as operator-written personal messages.

The Day 7 operator scorecard: Health score tier assignment at Day 7 produces the per-member intervention queue described in Table 3. For communities that run the intervention sequence manually, the Day 7 scorecard — a simple list of members in each tier with the recommended action — is the practical artifact the operator reviews each week. See the onboarding sequence reference card for the full Day 7 scorecard format (four activation gates per member: activation, specificity, connection, value) and the member retention reference card for the monthly version of this review at the cohort level.

Table 4 — Health score calculation method by community size and automation level

The health score is computable from Slack data using three different methods depending on community size and operator tooling. The correct starting method is always the manual Slack export approach: it requires no additional tooling, produces the same four-tier assignment outcome as the automated approach, and forces the operator to look at individual member data directly — which builds the pattern-recognition instinct needed to evaluate whether the automated version (once implemented) is behaving correctly. Operators who skip the manual phase and go directly to automated tooling often find that the health score is running correctly but the tier thresholds are miscalibrated for their specific community’s channel structure, because they never developed the baseline intuition from reviewing individual member data manually.

Method	Setup time	Weekly review time	Data required	Accuracy vs. automated	Scalability	Best for
Manual Slack export	0 setup — available immediately to any workspace admin	20–40 min per weekly cohort review (scales linearly with cohort size; 5 new members = ~5 min; 30 new members = ~30–40 min)	Slack workspace export (JSON) downloaded from Workspace Settings → Import/Export Data. Review the `channels/` folder for intro channel message history and the `users/` and `channels/` folders for per-member message counts. Available to paid Slack tiers (Pro, Business+, Enterprise).	Equivalent accuracy to automated for communities where the operator can review each new member’s post history directly. Marginally lower accuracy for communities with high message volume where manual review may miss threads in busy channels.	Not scalable beyond 50 new members per month. At 30 members per week (120 per month), the manual review consumes 4–5 hours per week, which exceeds the ROI threshold for a one-person operator.	Communities with <50 new members per month. Starting point for any community regardless of size, to validate tier thresholds before automating.
Semi-automated (Slack API + spreadsheet)	2–4 hours of initial setup; requires basic familiarity with API key generation and spreadsheet formula construction (or a developer hour). Uses Slack API `conversations.history` and `users.list` endpoints. Free Slack API tier supports this use case.	5–15 min per weekly cohort review (spreadsheet auto-populates signal counts; operator reviews tier assignments and queues interventions)	Slack API key (generated from api.slack.com; requires workspace admin or app scope); a spreadsheet (Google Sheets or Excel) with API calls pulling message counts per channel per member, structured to apply the signal weight formula and produce the four-tier assignment automatically.	High. Pulls directly from Slack API without export lag; captures real-time data at the time of review.	Scales to 200–300 new members per month with the weekly spreadsheet review workflow. Above 200 members per month, the API rate limits on the free Slack API tier may require batching the weekly pull across multiple sessions.	Communities with 50–200 new members per month where the operator or a team member has technical comfort with API setup and spreadsheet formulas. The correct middle step between manual review and purpose-built automated tooling.
Automated (purpose-built onboarding tooling)	One-time Slack OAuth install (typically <5 minutes). No spreadsheet construction required. Signal capture begins at the next member join event after install.	<5 min per weekly review (health score dashboard shows tier distribution by cohort; intervention queue pre-populated; Day 3 conditional nudge automated)	Slack OAuth install with appropriate event subscription scopes (`message.channels`, `member_joined_channel`). No workspace export or API key management required. Signal capture is event-driven at join, not batch-pulled weekly.	Highest accuracy. Real-time signal capture means no lag between the member action and score update; no risk of export timing errors affecting tier assignment.	Scales to any community size. Per-member computational cost is negligible; the practical limit is the operator’s capacity to review and send the Activated, Healthy, and Churned personal DMs (the three interventions that require operator personalisation).	Communities with >100 new members per month, or communities where the operator wants the Day 3 conditional nudge automated consistently without manual trigger review. Also appropriate for smaller communities where the operator prefers a dashboard view to manual data review.

The transition point from manual to semi-automated is approximately 50 new members per month — the level at which the manual review time (roughly 40–50 minutes per week) starts to exceed the operator’s capacity to also run the four-tier intervention queue. The transition from semi-automated to purpose-built tooling is approximately 150–200 new members per month — the level at which the semi-automated spreadsheet pull approaches Slack API rate limits and the intervention queue (particularly the At-Risk Day 3 nudge, which is the one intervention with a time-sensitive 72-hour window) benefits from event-driven automation rather than weekly batch review.

The recalibration check: Health score tier thresholds should be rechecked every 3 months against actual month-3 renewal data. If your Activated-tier members are renewing at below 65% (well below the 72–88% benchmark), the Activated threshold (currently ≥70 points) may be set too low for your community’s channel architecture. If your Healthy-tier members are renewing above 80%, the Healthy threshold may be set too conservatively and some Healthy-tier members could be reclassified as Activated for intervention purposes. The benchmark renewal rates in Table 5 are baselines; your community’s actual renewal rates by tier are the authoritative calibration data once you have 3+ months of member cohort history.

Table 5 — 90-day renewal rate benchmarks by health score tier and community price tier

The renewal rate differentials between health score tiers are remarkably consistent across community price tiers. The absolute renewal rates vary — higher-priced communities renew at higher absolute rates because higher price correlates with higher self-selection motivation and higher perceived cost of cancellation — but the relative gap between Activated and Churned members within each price tier remains approximately 45–55 percentage points regardless of price level. This consistency validates the health score’s price-agnostic design: the behavioral signals that predict renewal are the same for a $49/mo community as for a $300+/mo community, because the mechanism (value exchange events in week one producing a felt sense of community belonging) is the same. What changes with price is the baseline renewal probability at each tier, not the tier’s relative position within the distribution.

Price tier	Activated (≥70 pts)	Healthy (40–69 pts)	At-Risk (15–39 pts)	Churned (0–14 pts)	Activated–Churned gap	Notes
Under $50/mo (entry-level paid communities, low-ticket recurring)	65–78%	52–65%	28–42%	14–25%	~47pp	Lower absolute rates reflect lower perceived commitment at the price point — cancelling a $29/mo subscription carries lower psychological cost than cancelling a $299/mo subscription. Entry-level communities benefit disproportionately from the Day 0 DM because the activation event (intro post) is the only week-one mechanism that creates a felt sense of investment that offsets the low-commitment price signal.
$50–$149/mo (mid-tier paid communities; primary SMB market)	72–85%	58–72%	32–48%	18–32%	~50pp	The benchmark range for most paid Slack communities in the Foothold ICP ($49–$199/mo). The Activated–Churned gap of ~50pp is the most reliable benchmark for evaluating whether an onboarding improvement is working: if you move 10% of your new-member cohort from Churned to Activated, you add approximately 5pp to your cohort’s overall 90-day renewal rate. At $99/mo with 20 new members per month, that’s approximately $11,880 in additional annual LTV per cohort.
$150–$299/mo (upper-mid-tier paid communities, professional / executive cohorts)	76–90%	62–76%	38–54%	22–36%	~51pp	Higher absolute rates reflect stronger self-selection: members who pay $200+/mo for a community have made a considered purchase decision and are more motivated to extract value. The At-Risk and Churned rates are still substantial at this price tier because motivated buyers can still experience the same onboarding failure mode (unclear intro path, overwhelming channel list, goal mismatch) that affects lower-priced community members.
$300+/mo (premium paid communities, high-ticket B2B cohorts)	80–92%	68–80%	44–58%	28–42%	~49pp	The Activated rate at premium price tier reaches 80–92% partly because of price-point self-selection and partly because premium communities typically have lower member volume (100–500 members) that allows the operator to run more personalised onboarding. The Churned rate remains non-trivial at 28–42% because premium-ticket buyers have higher expectations that are more easily unmet in week one. The Churned personal DM is particularly high-value at this tier: recovering one Churned premium member represents $3,600–$10,000+ in annual LTV.

The renewal rate data in Table 5 is the business case for the health score itself. The 45–55pp gap between Activated and Churned members at every price tier means that moving members from Churned to Activated tier is the highest-ROI investment any paid community operator can make in their retention system. The question is not whether to build a health score, but how much operational effort the operator can sustain to capture the signals, assign the tiers, and execute the four interventions. At communities under 50 new members per month, the answer is 20–40 minutes per week of manual review and 4 personalised DMs per week at typical tier distributions. The expected LTV return from that 20–40 minutes at $99/mo is approximately $800–$1,400 per month in retained subscription revenue that would otherwise cancel at month 3.

The health score and churn by tenure: The month-3 renewal rates in Table 5 represent the most consequential tenure gate in the paid community member lifecycle. Members who reach month 3 as Activated or Healthy tier members renew at substantially higher rates at month 6 and month 12 as well, because the community relationship established in week one compounds: peer interactions in week 1 produce returning contributors in months 2–3, which produces a multi-year member whose LTV is 3–5× the month-3-only retainer. Members who reach month 3 as Churned-tier members — who have been passively present but not contributed — typically cancel at their first billing review date rather than their second, making month 3 the gate that determines the difference between a 3-month subscriber (LTV: 3× monthly price) and a 24-month subscriber (LTV: 24× monthly price).

Putting the five tables together: the weekly health score review routine

The five decision tables in this reference card are designed to be used together as a weekly review routine. At Day 7 for each new-member cohort, the operator calculates the three signal values per member (Table 1), assigns each member to a tier using the score and threshold table (Table 2), queues the corresponding intervention for each member (Table 3), and runs the review using whichever calculation method matches the community’s scale (Table 4). Table 5 provides the baseline renewal rate expectations for each tier at the current price point, which the operator compares against actual month-3 renewal rates every quarter to validate whether the tier thresholds are correctly calibrated.

The full routine for a community with 20 new members per week takes approximately 25–35 minutes using the manual Slack export method: 15–20 minutes of signal calculation (reviewing intro post completions, peer reply counts, and non-intro channel activity for 20 members), 5 minutes of tier assignment and intervention queue setup, and 5–10 minutes of writing the four personalised DMs (Activated recognition, Healthy goal-check, At-Risk Day 3 nudge if not already automated, Churned personal DM). At $99/mo, the expected return from this weekly 25–35 minute investment is approximately $600–$900 per month in incremental LTV from the intervention conversions across all four tiers.

The health score is not a replacement for the three-touch welcome sequence — it is a diagnostic layer on top of it. The welcome sequence (Day 0 DM, Day 3 conditional nudge, Day 7 scorecard) drives the behavioral signal inputs that the health score measures. A community with a well-functioning welcome sequence will see fewer Churned-tier members and more Activated-tier members at Day 7 because the sequence has already executed the At-Risk intervention (the Day 3 conditional nudge) and improved the peer interaction signal (by asking a goal-track question in the Day 0 DM that generates replies). The health score quantifies the outcome of the welcome sequence and identifies the members who have not yet responded to the sequence’s automated touchpoints — the Churned tier — who need a personal operator DM that the automated sequence cannot send.

Related reference cards & posts

Paid community welcome sequence reference card — the three-touch sequence (Day 0, Day 3, Day 7) that produces the behavioral signals the health score measures.
Paid community member activation rate reference card — cohort-level activation benchmarks and how to calculate the real number from Slack workspace data.
Slack community health metrics reference card — six dashboard metrics (weekly active poster rate, day-3 nudge response rate, etc.) that complement the individual health score at the cohort level.
Paid community member retention reference card — the monthly cohort review framework and five-lever retention system for operators past the first-month gate.
Paid community member churn by tenure — churn rate benchmarks by month-0 through month-24 tenure, and why week-one activation is the month-3 churn predictor.
Why your paid community member activation rate is probably wrong — the blog companion on how login metrics systematically overstate activation by 25–40pp and why the gap matters.
Take the 2-minute Onboarding Health Check — five questions, a 0–50 score, and the top three onboarding fixes for your community.