What are the five metrics every paid community should track?

The five metrics that predict month-3 retention for paid Slack communities are: (1) Activation rate — the percentage of new members who complete all three Day 0–7 activation steps (introduce in #introductions, state their primary goal, subscribe to at least one goal-track channel). Healthy range: 35–50% with structured onboarding; below 25% is a structural problem. (2) First-week post rate — the percentage of new members who publish at least one original post (not a reply) in any channel during their first 7 days. Healthy range: 45–60% with Day 0 onboarding DM; below 40% triggers Day 3 nudge audit. (3) 30-day contribution rate — the percentage of new members who post at least once in any channel during their first 30 days (not counting #introductions posts). Healthy range: 55–70%; below 50% signals that Day 7 activation did not convert to habit. (4) Monthly engagement rate — the percentage of all subscribed members who post in any channel in any rolling 30-day window, measured on all members subscribed for 30 or more days. Healthy range: 40–60%; below 30% indicates a lurker-heavy community where the active minority is not sufficient to generate social proof for month-3 renewal. (5) Cohort month-3 retention — the percentage of a join cohort (all members who joined in a specific calendar month) still subscribed at the end of month 3. Healthy: 70%+; at-risk: 55–70%; structural problem: below 55%. This is the single lagging indicator that all four leading metrics predict. The three metrics most operators track instead — total member count, daily active users, and message volume — do not predict month-3 renewal because all three can be inflated without a single churned member being retained.

What is a healthy activation rate for a paid Slack community?

A healthy activation rate for a paid Slack community with structured onboarding is 35–50%. Activation rate is calculated as the number of new members who complete all three steps of the Day 0–7 onboarding sequence divided by the number of new members who joined in the same calendar period, expressed as a percentage. The three activation steps are: (1) post an introduction in #introductions or the designated intro channel; (2) state a specific goal or use case (in the intro post, in a goal-selection prompt in the Day 0 DM, or in a dedicated channel); (3) join at least one goal-track opt-in channel based on their stated goal. At-risk threshold: 25–35%. At this range, the Day 3 nudge copy should be audited — most commonly the nudge is generic (“check out the community”) rather than goal-specific (“you said X in your intro — here are the three channels where that conversation is happening”). Structural-problem threshold: below 25%. At this level, the problem is typically pre-Day-3 abandonment — the Day 0 DM either was not delivered (operator has not configured a Slack app to send it), was too long to act on in a single session, or lacked a concrete first step that a new member could complete in under 3 minutes. The most common failure cause at below 25%: the first step in the Day 0 DM asks for a 3–5 sentence introduction in a public channel, which carries social risk for new members. Members who have not yet established credibility in the community avoid posting publicly until they have read enough threads to understand the community norms. Fix: make the first step of the Day 0 DM a private, low-stakes action (select your goals from a list, or reply to the DM with a one-line answer to a specific question) before asking for a public introduction.

How do you calculate monthly engagement rate for a paid community?

Monthly engagement rate for a paid community is calculated as: (unique members who posted at least once in a rolling 30-day window) ÷ (total members subscribed for 30 or more days) × 100. The denominator deliberately excludes members in their first 30 days because new members have a structurally different posting pattern during activation (higher in week one if the onboarding flow works, then variable in weeks 2–4 as habits form). Including them in the denominator would create a noisy metric that moves with cohort-join-rate variability rather than with community health. Healthy range: 40–60%. Below 30% indicates a lurker-heavy community. The lurker-heavy pattern has a specific failure mode: the 30% who post generate visible social proof, but the 70% who lurk do not form peer relationships, do not have a post-history to reference when they consider cancelling, and cancel at month 3 at rates 25–35 percentage points higher than the posting minority. A monthly engagement rate below 30% typically does not improve by asking lurkers to post — generic calls to participate are ignored. The root cause is that lurkers do not have a clear, specific, low-stakes post type available to them. Fix: add structured contribution formats that lower the friction of a first post in months 2–3 (a monthly “what are you working on” thread, a member-spotlight question prompt, a wins-and-lessons format that does not require a long write-up). Measure monthly engagement rate separately for the cohort that joined in the past 90 days versus the cohort that has been subscribed for 91–365 days — the two populations have different healthy ranges (new cohort 45–65%; established cohort 35–55%).

What does cohort month-3 retention below 55% indicate?

Cohort month-3 retention below 55% indicates structural churn: the community is losing 45% or more of new members before their fourth payment. At this level, the community cannot compound its subscriber base — every cohort of new members produces net-negative growth after netting out month-3 cancellations against new joins. The cause is almost always in week one of the member’s tenure, not in month three. Members who cancel at month 3 almost universally show the same week-one fingerprint: they never completed the activation sequence (no intro post, no goal-track channel joined, no peer connection formed in the first 7 days). By month 3, their mental model of the community is formed entirely from passive reading, not participation. They can leave without social loss because they have no peer relationships to lose. The intervention point is always week one of each new cohort — not month three when the cancellation is visible. At below 55% cohort month-3 retention: (1) audit the Day 0 DM delivery rate (confirm the Slack app is sending it to every new joiner and not just new channel-type joins); (2) check the activation rate for the same cohort — if activation rate is below 25%, the Day 0 DM is not converting to action; (3) check the first-week post rate — if below 40%, the Day 3 nudge is either not being sent or not containing a specific, actionable recommendation; (4) check whether any member in the sub-55% cohort formed at least one peer connection (replied to another member’s non-operator thread) in their first 30 days — peer connection is the single strongest behavioral predictor of month-3 renewal across paid Slack communities in the 200–2,000 member range.

Metrics Reference Card

Paid community metrics dashboard — five-metric benchmark table, calculation reference, vanity metric failure modes, weekly review checklist, and cohort retention reference card

This page is a structured reference card for paid Slack community operators who want the benchmark numbers, calculation formulas, and weekly review steps in scannable table and checklist form. It covers: a five-metric benchmark table with healthy range, at-risk threshold, structural-problem threshold, and primary intervention per metric; a calculation reference table with numerator, denominator, data source, and recommended review frequency for each metric; a vanity metric failure-modes table explaining what each commonly tracked number measures, what it does not predict, and why it is excluded from the five; a five-step weekly review checklist with 3-minute time budgets and decision thresholds; and a cohort retention reference table with expected still-subscribed and still-active rates by month for structured versus unstructured onboarding, plus intervention type if below floor. For the strategic reasoning behind why these five metrics were selected over more familiar alternatives, including the vanity-metric trap, the social-proof mechanism that drives month-3 renewal, and the difference between still-subscribed and still-active as a leading indicator of silent churn, see the companion post: The 5 paid community metrics that predict month-3 retention. This card is for the operator who understands the “why” and needs the benchmarks and thresholds in quick-reference form.

TL;DR

Track five metrics: activation rate (healthy 35–50%), first-week post rate (healthy 45–60%), 30-day contribution rate (healthy 55–70%), monthly engagement rate (healthy 40–60%), cohort month-3 retention (healthy 70%+). Review weekly in 15 minutes. Any metric below its at-risk threshold triggers the same-week intervention in the benchmark table below. Below the structural-problem threshold, escalate to an onboarding audit before the next cohort joins. Do not track total member count, DAU, or message volume — none of the three predict month-3 renewal.

Five-metric benchmark table

The table below defines the five metrics that predict month-3 retention for paid Slack communities, with the healthy range, at-risk threshold, structural-problem threshold, and the primary intervention for each. “Healthy” means a community with structured onboarding (Day 0 DM + Day 3 nudge + Day 7 scorecard). Without structured onboarding, expect all five numbers to run 15–25 percentage points below the healthy range. “At-risk” means the metric signals a problem that will appear in cohort month-3 retention in 6–10 weeks if the intervention is not applied. “Structural problem” means the metric reflects a root-cause issue in the onboarding sequence itself, not in the content or programming layer — and cannot be fixed by adding content or events without first fixing the flow.

Metric	Healthy range	At-risk threshold	Structural-problem threshold	Primary intervention
Activation rate	35–50% — percentage of new members who complete all three Day 0–7 activation steps: intro post in #introductions, stated goal in a DM prompt or intake form, and at least one goal-track opt-in channel joined.	25–35% — Day 3 nudge copy audit required. Most likely cause: nudge is generic (“check out the community”) instead of goal-specific (“you said X — here are the three channels where that conversation happens”). Also check: Day 3 DM delivery rate, especially if community uses Workflow Builder rather than a Slack app (Workflow Builder sends fail silently if the member has not engaged with the workspace in the preceding 7 days).	Below 25% — Day 0 DM not reaching members, or first step too high-friction. Most common cause: first activation step asks for a 3–5 sentence public introduction before the member has read enough threads to understand community norms. Fix: replace the first public step with a private low-stakes action (select your goals from a 4-option list; reply to the DM with a one-line answer). Also check: is the Slack app sending a DM to every workspace join, or only to members who join a specific channel?	Audit the Day 0 DM copy and delivery log. Confirm the trigger fires on workspace join, not channel join. Reduce friction on the first step (private before public). Verify the Day 3 nudge includes the specific, unfinished step from the Day 0 sequence, not a generic re-invite.
First-week post rate	45–60% — percentage of new members who publish at least one original post (not a reply) in any channel during their first 7 days. Members who post in week one renew at month 3 at 3–4× the rate of members who never post. First-week post rate is a stronger retention predictor than activation rate alone because it measures actual content contribution, not just checklist completion.	40–45% — the Day 3 nudge is sending but the nudged action is not producing posts. Most likely cause: the specific step the nudge recommends requires too much social risk for a member who has not yet established credibility in the community. Check: does the nudge recommend posting in a channel where new member posts are likely to receive replies within 24 hours? A new member’s first post that receives no response produces a worse activation outcome than not posting at all.	Below 40% — the community lacks a clear, low-social-risk post type for new members in week one. Generic “introduce yourself” in a high-volume channel carries more perceived risk (my post will be buried, nobody will reply) than a structured prompt format (this week’s “what are you working on” thread). Fix: add a pinned weekly prompt thread to the primary content channel as a designated landing zone for first posts from new members. Notify new members in the Day 3 nudge about the current week’s prompt thread specifically.	Audit which channel the Day 3 nudge recommends for a first post. Confirm that channel has a visible, recent thread where new posts receive replies within 24 hours. Consider adding a weekly structured prompt thread. Check whether posts from members in their first 7 days are receiving replies from established members (reply rate to new-member first posts below 60% is the primary cause of sub-40% first-week post rate).
30-day contribution rate	55–70% — percentage of new members who post at least once in any channel during their first 30 days, excluding their #introductions post (which is activation-sequence-driven and does not reflect spontaneous habit). This is the conversion metric from “activated in week one” to “has an established posting habit.”	50–55% — the Day 7 scorecard is marking members as activated without a sufficient hook for continued participation. Most likely cause: the Day 7 touch is informational (a summary of the week’s best threads, or a reminder about upcoming events) rather than relational (a direct question about what the member is working on, with an invitation to post a specific update). Informational Day 7 touches do not extend the activation window into week 2.	Below 50% — week-one activation did not convert to social proof sufficient to sustain participation. Root cause: new members activated (completed the checklist) but did not form a peer connection (did not receive a meaningful reply from another member who was not the operator or community manager) in their first 7 days. Fix: audit week-one peer reply rates. If a new member’s first post in any channel received zero replies from non-operator members, the activation is statistically equivalent to not activating for the purpose of month-3 renewal.	Review the Day 7 scorecard copy. Replace informational content with a direct question that invites a specific post in weeks 2–3 (“You mentioned X in your intro — what’s the one thing you’re trying to figure out about X right now?”). Audit whether non-operator members are replying to new-member first posts within 24 hours. If not, establish a peer-welcome rotation with 5 established members who commit to replying to one new-member post per week.
Monthly engagement rate	40–60% — percentage of all members subscribed for 30 or more days who post at least once in any channel in any rolling 30-day window. Measures community-wide participation density, not just new-member activation. A healthy monthly engagement rate means the active minority is large enough to generate the social proof (visible recent activity, reply rates, thread depth) that makes month-3 renewal feel automatic rather than a deliberate re-evaluation. Below 30%: the community looks quiet to month-3 renewal decision-makers.	30–40% — the community has a lurker-heavy pattern emerging. Members who joined 2–4 months ago are reading but not contributing. Most likely cause: there is no structured contribution format for members who are not ready to write a long original post but would participate if there were a lower-friction format. At this range, check whether the 30–40% who post are producing high-reply threads or primarily posting announcements and questions that receive no engagement — low-reply-rate posting at 40% engagement looks like dead activity to a lurker evaluating whether to renew.	Below 30% — the lurker majority is not forming peer relationships and will cancel at month 3 at 25–35 percentage points higher rates than the posting minority. Generic calls to participate do not fix this. Fix: add structured contribution formats that lower the friction of a first post in months 2–3: a monthly “what are you working on” thread, a member spotlight question format, a wins-and-lessons prompt that does not require a long write-up. Measure monthly engagement rate separately for members 0–90 days and 91–365 days — if the 91–365-day cohort is below 25%, the community has a retention cliff at month 3 that is not recoverable without structural programming changes.	Add at least one low-friction structured contribution format per month (a thread prompt with a question that can be answered in 1–2 sentences). Notify members in months 2–3 when their specific expertise area is mentioned in a thread (this produces replies from lurkers who feel qualified to answer a specific question). Review whether high-engagement posts are from a concentrated group of 5–10 members — if yes, the community has a social-proof dependency on a small group whose departure would collapse visible activity.
Cohort month-3 retention	70%+ — percentage of a join cohort (all members who joined in a specific calendar month) who are still subscribed at the end of month 3. This is the single lagging indicator that all four leading metrics predict. A community with 70%+ cohort month-3 retention compounds its subscriber base: new cohorts add more members than old cohorts cancel, producing net positive subscriber growth every month.	55–70% — the community retains a majority but loses more than it should. At 60% month-3 retention with 30 new members per month, the community grows by 18 net new members from each cohort at month 3 — but a 70% cohort adds 21 net new members, a compounding difference that produces a 15–20% larger community at month 12 without changing acquisition rate. Check week-one post rate for the same cohort: if week-one post rate is below 45%, the gap between healthy and at-risk retention is explained by the first-post conversion failure.	Below 55% — structural churn. The community cannot compound its subscriber base at this retention rate. At below 50%, every month you operate is producing a net-negative subscriber balance even if you are adding new members. The intervention point is always week one of the new cohort, not month three when the cancellation is visible. Same-week action: audit the Day 0 DM delivery rate, confirm activation rate for the same cohort, and check whether any sub-55% cohort member formed a peer connection (replied to another member’s non-operator thread) in their first 30 days.	Trace the sub-55% cohort’s week-one fingerprint: what was their activation rate, first-week post rate, and peer-reply rate? The cancellation at month 3 was determined by week-one behavior. Fix the earliest-failing metric for the next incoming cohort — do not add programming or events for month-2 if the week-one DM sequence has not been fixed first. Peer connection (at least one reply received from a non-operator member in the first 30 days) is the single strongest behavioral predictor of month-3 renewal — confirm this is happening for every member who completes activation.

Leading vs. lagging order. Activation rate, first-week post rate, and 30-day contribution rate are leading indicators: they are measurable at week 1, 1, and 30 of a member’s tenure respectively — well before month 3. Monthly engagement rate is concurrent: it is measured continuously and reflects the current state of community health across all active cohorts. Cohort month-3 retention is lagging: it is only measurable 90 days after a cohort joins. Use the four leading and concurrent metrics to predict and intervene; use cohort month-3 retention to confirm the intervention worked.

Calculation reference table

The table below defines the numerator, denominator, data source, and recommended review frequency for each of the five metrics. All five can be calculated from two sources available without additional tooling: Slack member data (Slack → Settings & administration → Manage members) and your payment processor’s subscriber list (Stripe, Memberstack, or equivalent). The calculation for activation rate requires a record of which members completed which activation steps — this data does not exist in Slack by default and must be tracked in a spreadsheet, a Slack bot log, or a tool like Foothold. The remaining four metrics can be approximated from Slack’s member activity export alone.

Metric	Numerator	Denominator	Data source	Review frequency
Activation rate	Count of new members who completed all 3 activation steps (intro post in #introductions + goal stated in DM or intake form + at least 1 goal-track opt-in channel joined) within 7 days of their workspace join date.	Count of all new members who joined the workspace in the same 7-day window, including members who have not yet reached day 7 (use the rolling-7-day count from 7+ days ago to avoid numerator lag).	Activation step completion: Slack bot log (if using Foothold or equivalent), or spreadsheet manually updated from DM history. Join date: Slack member export or Slack admin “Joined” column. Goal-track channel join: Slack channel member list export.	Weekly. Calculate for members whose 7-day window closed in the past 7 days. Do not combine cohorts — track per join-week so you can identify step-drop patterns tied to specific onboarding copy changes.
First-week post rate	Count of new members (joined in a specific 7-day window, now at 7+ days tenure) who published at least one original post (thread-starter, not a reply) in any channel within their first 7 days. Exclude #introductions posts — the activation sequence forces an intro post, so it does not reflect spontaneous first-week contribution.	Count of all new members who joined in the same 7-day window and have now reached 7+ days of tenure.	Slack member activity export (downloadable from Slack → Settings & administration → Manage members → Export). The “Messages posted” column gives total post count; to get original posts vs. replies, you need channel-level message data or a Slack bot log. For manual calculation: search each channel for posts from new members in their join week.	Weekly, aligned to the same 7-day cohort window as activation rate. The two metrics should always be reviewed together — a high activation rate with a low first-week post rate means members are completing the checklist without internalizing the “post here” action.
30-day contribution rate	Count of new members (joined in a specific calendar month, now at 30+ days tenure) who posted at least once in any channel in their first 30 days, excluding their #introductions post from the count (the intro post is forced by the activation sequence; a second post in any channel reflects a spontaneous return to the community).	Count of all members who joined in the same calendar month and have now reached 30+ days of tenure.	Slack member activity export (30-day post count by member). For the introductions-exclusion rule: subtract 1 from the post count of any member whose only post was in #introductions. Any member with 2+ posts in 30 days automatically passes regardless of channel distribution.	Monthly, for the cohort that joined 30+ days ago. The 30-day contribution rate for a cohort is a fixed number once all members reach 30 days — it does not change with additional time. Track it as a per-cohort metric and compare month over month to identify whether onboarding changes are producing improvement.
Monthly engagement rate	Count of unique members (subscribed for 30 or more days as of the measurement date) who posted at least once in any channel in the rolling 30-day window ending on the measurement date.	Count of all members subscribed for 30 or more days as of the measurement date, regardless of activity level. Do not exclude inactive members from the denominator — they are the lurkers whose non-participation is the signal you are measuring.	Slack member activity export (monthly active posters count) plus payment processor subscriber list (to confirm active subscription status). The Slack admin panel “Monthly active members” figure is not the same as monthly engagement rate — Slack counts a “monthly active member” as anyone who views a channel, including pure lurkers. Use the “Messages posted” count from the member export instead.	Weekly, using a rolling 30-day window. Track as a trend, not a point-in-time number. A community with a monthly engagement rate that is flat at 40% is healthier than one that spikes to 50% in event weeks and drops to 25% in non-event weeks — the variance signals event-dependency rather than sustained habit.
Cohort month-3 retention	Count of members who joined in a specific calendar month and are still subscribed (not cancelled, not refunded, not paused) at the end of month 3 (90 days after the last day of their join month).	Count of all members who joined in the same calendar month. Do not use current subscriber count as the denominator — new joins in the intervening months inflate the denominator and make retention look higher than it is. The denominator is fixed at the join-month count.	Payment processor subscriber list with join date and cancellation date. Filter: members whose join date falls within the target calendar month. Count: those still active at day 90 from their join date. If your payment processor does not export cancellation dates, use Stripe’s customer data export (Stripe Dashboard → Customers → Export) with “created” and “subscription_status” columns.	Monthly. Once a cohort reaches month 3 (90 days from join), its retention rate is fixed. Track the trend across the past 6 cohorts to identify improvement (if you changed something in the onboarding sequence) or regression (if cohort quality or community health declined). Do not average across cohorts — keep them separate to preserve signal from individual cohort variation.

Vanity metric failure modes table

The three metrics most paid community operators track instead of the five above — total member count, daily active users, and message volume — are described as “vanity metrics” because they can all increase while month-3 retention deteriorates. The table below explains what each vanity metric actually measures, what it does not predict, the most common inflation mechanism (how the number grows without the underlying behavior improving), and why each is excluded from the five-metric dashboard.

Vanity metric	What it measures	What it does NOT predict	Common inflation mechanism	Why excluded
Total member count	The cumulative count of all members who have ever joined the workspace and have not yet cancelled, regardless of their activity level, posting behavior, or renewal intent. A 500-member community by this measure may have 300 members who have never posted and will cancel at month 3.	Month-3 renewal rate. New member adds outpacing cancellations makes total member count grow even when cohort month-3 retention is below 55%. The metric feels like growth; the underlying subscriber base is churning faster than it appears. Total member count also does not distinguish between paying members (subscribed) and lapsed members (trial-ended or grace-period) who have not yet been removed from the workspace.	A strong acquisition month adds 30 new members. Total count rises by 30 even if 20 of those members cancel before month 3. Operators celebrate the acquisition month; the retention failure appears 90 days later as a separate, unconnected event. The temporal gap between the acquisition win and the retention loss makes total member count a misleading real-time signal.	Total member count is a cumulative counting metric, not a rate metric. It cannot be lower than last month’s number unless cancellations exceed joins. It carries no information about whether the existing subscriber base will renew — a 2,000-member community with 40% cohort month-3 retention is shrinking; a 200-member community with 80% cohort month-3 retention is compounding. Use cohort month-3 retention instead.
Daily active users (DAU)	The count of unique members who view at least one message in any channel on a given day. Slack’s definition of “active” is a message view, not a post, reply, or reaction. A member who opens Slack, reads the unread badge on a channel, and closes the app counts as a DAU. Pure lurkers inflate DAU without contributing to the social proof that retains them.	Month-3 renewal rate. DAU and cohort month-3 retention are weakly correlated in paid communities because Slack sends notification badges that draw members back to the workspace without triggering any participation behavior. Members who are listed as DAU but never post are the highest-churn segment in paid Slack communities — they open the app enough to feel they are getting their money’s worth, but their lack of peer relationships means there is no social cost to cancelling.	Onboarding DM delivery triggers a notification that brings members back to Slack. Each of the Day 0, Day 3, and Day 7 touches produces a DAU spike for the cohort even if no member responds to any of the three messages. A community with a high-volume onboarding sequence looks more “active” by DAU than a community with a low-volume sequence, regardless of whether the high-volume sequence is producing posts or peer connections.	DAU measures visits, not participation. Monthly engagement rate measures participation (posts, which require reading, forming an opinion, and accepting the social risk of being visible). The conversion from visit to post is where retention is won or lost. Tracking DAU without tracking the visit-to-post conversion rate produces a high-confidence false signal: the community looks busy while the underlying participation rate that predicts renewal is invisible.
Message volume	The total count of messages posted in all channels over a given period, including replies, reactions counted as message-equivalents (in some export formats), automated bot messages, operator announcements, and community manager posts. In a 500-member community, 80% of message volume in a typical week may be generated by 10–20% of members plus the operator account.	Month-3 renewal rate. Message volume can be high in a community where 10 highly active members generate 90% of posts and 80% of members are passive readers. The passive majority does not renew at month 3 at the same rate as the posting minority, but message volume does not distinguish between the two. A community with 5,000 messages per month from 15 members and 490 lurkers has the same message volume as one where 200 members each post 25 times — but radically different renewal rates.	The operator posts a controversial topic or hosts an asynchronous Q&A event that generates 300 replies in 24 hours. Message volume spikes. The spike looks like community health in a dashboard but is event-driven, not sustained habit. When the event ends, message volume returns to baseline. Operators who read message volume as a health indicator over-weight event weeks and under-weight the non-event steady state, which is the actual health signal.	Message volume is a count metric that conflates operator-generated content, event-driven spikes, and the contributions of a small highly-active minority. Monthly engagement rate — which counts unique members who post in a 30-day window, not total messages — is a better indicator of participation breadth. A community where 45% of members post each month (potentially one post each) is healthier for renewal purposes than one where 10% post an average of 20 times each, even if the latter produces 2× the message volume.

The common thread across all three vanity metrics: each one can be improved without any improvement in the behaviors that predict month-3 renewal (peer connection formation, goal-track channel participation, spontaneous return posts in months 2–3). Total member count rises with acquisition. DAU rises with notification volume. Message volume rises with operator activity. None of the three requires a single lurker to post, connect with a peer, or form a reason to stay.

Five-step weekly review checklist

The checklist below gives a five-step structure for running the weekly community metrics review in 15 minutes or fewer. Each step has a 3-minute time budget, a specific metric to check, a decision threshold that triggers action, and the action to take if the metric is below threshold. Run this review on the same day each week. The five steps are ordered from leading (metrics you can act on immediately this week) to lagging (metrics that confirm last month’s interventions worked). For the full context behind why this review order matters — including why checking cohort retention before checking activation rate produces the wrong intervention sequencing — see the companion post: The 5 paid community metrics that predict month-3 retention.

Step	Time budget	What to check	Decision threshold	Action if below threshold
Step 1	3 minutes	Activation rate for the cohort whose 7-day window closed in the past 7 days. Count members whose join date was 7–14 days ago; count how many completed all 3 activation steps; divide. Compare to the prior week’s activation rate for the same calculation period to distinguish trend from noise.	Below 35%: in the at-risk band. Below 25%: structural problem, immediate action.	If 25–35%: read the Day 3 nudge copy sent to the non-activated members from this cohort. Is it goal-specific or generic? Replace generic nudge copy with the specific unfinished step from each member’s Day 0 DM before the next cohort hits Day 3. If below 25%: check Day 0 DM delivery log. If delivery rate is below 95%, fix the trigger condition first. If delivery rate is above 95%, reduce the friction of the first activation step before the next cohort’s Day 0.
Step 2	3 minutes	First-week post rate for the same cohort (join date 7–14 days ago). Count members who posted at least one original post (not an intro post) in any channel in their first 7 days. Compare to the prior week’s cohort.	Below 45%: at-risk band. Below 40%: structural problem.	If 40–45%: identify which channel the Day 3 nudge recommended for a first post. Check the reply rate on new-member first posts in that channel in the past 7 days. If reply rate is below 60%, notify 3 established members to reply to any new-member first post this week. If below 40%: add a pinned weekly prompt thread to the primary content channel before next Monday. Include a specific mention of this thread in the Day 3 nudge copy for the next cohort.
Step 3	3 minutes	30-day contribution rate for the cohort that joined 30–37 days ago. Count members who posted at least twice in any channel in their first 30 days (counting their intro post as post #1; a second post anywhere passes this check). This step also functions as a spot-check on whether the Day 7 scorecard produced any week-2 or week-3 posts.	Below 55%: at-risk band. Below 50%: structural problem.	If 50–55%: check whether the Day 7 scorecard for this cohort asked a specific question that could be answered with a one-post reply in a channel. If the Day 7 touch was informational only (a summary of the week’s threads), revise it to include a direct question before the next cohort reaches Day 7. If below 50%: check peer reply rate for this cohort’s week-one posts. If new-member first posts received zero replies from non-operator members, initiate a peer-welcome rotation with 5 established members immediately.
Step 4	3 minutes	Monthly engagement rate for the rolling 30-day window ending today. Count unique members (subscribed 30+ days) who posted in any channel in the past 30 days; divide by total members subscribed for 30+ days. Track as a 4-week trend — weekly fluctuation of ±3 percentage points is normal; a 4-week declining trend is a signal regardless of absolute level.	Below 40%: at-risk band. Below 30%: structural problem. Declining 4-week trend below 40%: treat as structural-problem regardless of absolute level.	If 30–40% or declining trend: check whether any structured contribution format (monthly prompt thread, wins-and-lessons format, member spotlight question) ran in the past 30 days. If not, launch one this week. If below 30%: identify the top 10 lurkers by tenure (subscribed 60–90 days, zero posts in 30 days). Send each a personal DM this week referencing a specific thread in their goal-track channel with a low-friction reply invitation (“I thought of you when I saw this thread — would you add your take?”).
Step 5	3 minutes	Cohort month-3 retention for the cohort that joined 90–97 days ago. Count how many of those members are still subscribed today. This is the confirmation step — if the prior four steps have been executed correctly over the past 12 weeks, cohort month-3 retention should be at or above 70%. Use this step to verify, not to intervene (intervention at month 3 is too late to change the renewal decision for the current cohort).	Below 70%: at-risk. Below 55%: structural problem. Use the sub-55% cohort’s week-one fingerprint (Steps 1–3 metrics for that same cohort, measured 12 weeks ago) to identify which leading metric failed first.	If 55–70%: trace back to the cohort’s week-one activation rate and first-week post rate (recorded 12 weeks ago). The metric that was below its at-risk threshold at week one is the root cause of the month-3 gap. Fix that specific metric for the current incoming cohort. If below 55%: run the same trace. Additionally, confirm that at least one intervention was applied at each of Steps 1–4 during the 12-week window for this cohort. If no intervention was applied at the point where the leading metric was below threshold, the sub-55% outcome was predictable and preventable with a same-week action.

Cohort retention reference table

The table below shows expected still-subscribed and still-active rates by month for two community types: structured onboarding (Day 0 DM + Day 3 nudge + Day 7 scorecard, delivered via Slack app) and unstructured onboarding (manual welcome DM or no onboarding automation). The “still-subscribed” rate is the raw retention count: how many of the original join-month cohort are still paying. The “still-active” rate is a subset: how many of the still-subscribed members posted in the past 30 days. The gap between still-subscribed and still-active represents the “silent subscriber” risk inventory — members who are paying but not participating, and who are evaluating whether to cancel in the next renewal cycle. An operator who only tracks still-subscribed misses the silent-subscriber cohort, which cancels at month 6 or month 12 at rates 20–30 percentage points higher than the still-active cohort. For the full Slack community health metrics reference covering all six operational numbers, see the linked page.

Month	Still-subscribed % (structured)	Still-subscribed % (unstructured)	Still-active % (structured)	Still-active % (unstructured)	Intervention type if below floor
Month 1	92–97%	80–90%	60–75%	30–45%	Verify Day 0 DM delivery. If still-active is below floor, audit first activation step friction. Month-1 cancellations are almost always caused by a failure to deliver the Day 0 DM or a Day 0 DM so high-friction that the member never reaches step one.
Month 2	82–90%	65–78%	50–65%	22–35%	If still-subscribed is at floor but still-active is below floor, the community has silent subscribers forming. Launch a structured contribution format (monthly prompt thread) and a personal DM to each silent subscriber that references their stated goal from their intro post. Personalization is required — generic re-engagement messages are ignored by silent subscribers.
Month 3	70–82%	50–65%	42–58%	18–28%	Month-3 is the primary renewal evaluation point for most paid communities. Members who are still-subscribed but not still-active at month 3 are in the highest-risk cancellation window. The intervention that most changes month-3 outcomes is a proactive operator DM — not a billing reminder, but a direct question about whether the member got what they came for. Members who receive a personal check-in at month 3 renew at 15–20pp higher rates than members who receive only an automated renewal email.
Month 4–5	65–78%	42–57%	38–53%	15–25%	Members who survived month-3 renewal are more stable but not immune. The month-4–5 window is where the “high contribution, low engagement” diagnostic is relevant: members who posted heavily in months 1–2 but dropped to zero in months 3–4 are re-evaluating. Check the monthly engagement rate for this cohort segment separately. A structured spotlight or contribution format that re-invites month-4–5 members specifically (not the whole community) is more effective than a community-wide call to participate.
Month 6	60–73%	35–50%	35–48%	12–22%	Month-6 is the secondary cancellation cluster, driven by the still-subscribed-but-not-still-active cohort that survived month 3 without re-engaging. Members who have not posted in 90 days and have been subscribed for 6 months are the highest-priority segment for a retention intervention at this point. A personal DM referencing a specific thread or outcome in their goal-track channel converts to re-engagement at 20–30% — approximately 3× the rate of a generic “we miss you” message.
Month 12	52–65%	25–40%	28–42%	8–18%	Month-12 is the annual renewal decision point. Members who are still-active at month 12 renew at 72–82% with no intervention. Members who are still-subscribed but not still-active at month 12 renew at 40–55% with a personal renewal conversation, or 15–25% with only an automated billing email. The month-11 personal renewal conversation — a direct DM asking about the member’s outcomes and whether annual pricing makes sense for their use case — is the highest-ROI retention intervention in the entire 12-month cohort arc. Run it for every member who is still-subscribed-but-not-still-active at month 11. Do not send it to still-active members — they will renew at 72%+ without it, and a renewal-focus DM at month 11 to an active member signals that you are worried about losing them, which plants doubt.

The silent subscriber gap. The difference between the still-subscribed rate and the still-active rate at any given month is the silent subscriber inventory — members who are paying but not participating. At month 6 in a structured-onboarding community, that gap is approximately 25 percentage points (60–73% still-subscribed, 35–48% still-active). In a community of 500 paying members, that gap represents 125 members who are one bad renewal email away from cancelling. These members cannot be recovered by content or events — only by a personal conversation that reconnects them to their original goal for joining.