Paid community moderation guide: the three-tier norm framework
Most paid community operators have no moderation framework. What they have instead is a reactive process: something surfaces, they handle it, they move on. The next thing that surfaces gets handled differently, because by then the context has changed, the operator’s judgment has evolved, and the same behaviour that was ignored last month produces a different response today. Members who observe this pattern — and members who pay for community membership observe it closely — reach a conclusion that the operator finds unjust and that is entirely accurate: the enforcement is inconsistent. The standards are whatever the operator decides they are on a given day, applied to the members the operator notices, in ways that feel arbitrary to everyone who was not directly involved in the specific incident that prompted the response.
Reactive moderation compounds over time in three ways. The inconsistency problem grows: every undocumented decision becomes an informal precedent that contradicts the next undocumented decision. By month six, the operator is operating on a set of informal rules that have been contradicted often enough that no one in the community can predict how a given situation will be handled. The operator overload problem grows: every dispute goes directly to the operator because there is no visible process members can consult, and there is no documented basis the operator can point to when explaining a decision. The precedent-creep problem grows: members who want to test the limits of what is acceptable have no documented standard to compare their behaviour against, so the informal rule becomes whatever the most recent intervention established.
The fix is not more rules. It is a different structure for thinking about which rules are needed, why each type of rule requires a different moderation response, and how the operator documents and applies consequences consistently. This guide covers the three-tier norm framework that structures moderation decisions, the four-step dispute resolution ladder that handles the situations the framework doesn’t prevent, the five properties that make community rules enforceable, and the five wrong patterns that erode member trust even among members who were never directly involved in a dispute.
1. Why reactive moderation fails paid communities specifically
Reactive moderation is worse in a paid community than in a free community for a structural reason: every member is paying. When a member in a free community is removed, they lose access to something that cost them nothing financially. When a member in a paid community is removed — or even receives an inconsistent moderation response that feels like unfair treatment — they experience that inconsistency as something that was done to them in a context where they paid to be there. The threshold for feeling treated arbitrarily is lower. The likelihood of mentioning that feeling to someone else is higher. And the referral behaviour of members who witness a moderation decision they don’t understand drops measurably, even among members who were not personally involved.
The inconsistency problem has a specific mechanism. When an operator handles a moderation situation without a documented framework, they make a judgment call based on the specifics of that moment. That judgment call is correct for that situation. But it is also now an informal precedent: the next member who encounters a similar situation expects a similar response. When the next response is different — because the operator is in a different context, is dealing with a different member relationship, or simply doesn’t remember the previous decision — the inconsistency is visible to every member who observed both situations. The community learns that the standard is the operator’s current mood rather than a principled position, and that learning changes how members interpret everything the operator does.
The operator overload problem is self-reinforcing. A community with no documented moderation framework produces every dispute at the operator level because there is no visible process for members to apply on their own. Members who observe norm violations don’t know whether to say something, ignore it, or flag it to the operator; the uncertainty produces inaction among members who might otherwise self-regulate and escalation to the operator for situations that a simple peer norm-naming could have resolved. The operator becomes the only moderation resource in the community, which is unsustainable at scale and produces slower responses than a framework that lets members understand and enforce norms themselves.
The paid community member churn by tenure guide covers the four windows when churn happens and why. Months two through three are particularly vulnerable to moderation-related churn: members who have completed onboarding and begun contributing are now close enough to the community culture to notice when enforcement feels inconsistent. A member who observes an operator making a decision they don’t understand — especially one that touches a community norm they hold — starts evaluating whether this is a community where they can trust the standards. That evaluation, happening during the period when the member is deciding whether to fully invest in the community or hold back, has a disproportionate effect on month-three retention.
2. The three-tier norm framework
Not all community norms require the same type of moderation response. The most common mistake operators make when they build a moderation framework is treating all norm violations as equivalent — either ignoring everything under a certain severity threshold or escalating everything above one. The three-tier framework distinguishes between three types of norms, each of which requires a different type of response when violated.
Community norms are the cultural values that govern how the community behaves as a whole: what kinds of sharing are valued, what the relationship to status and self-promotion is, whether critique is invited or defaults to encouragement, how the community handles failure and setback posts. Community norms are rarely written down explicitly. They exist in the patterns of what gets replied to enthusiastically, what gets quiet acknowledgement, and what produces no response at all. When a community norm is violated — a member posts in a way that conflicts with the underlying culture even if it doesn’t break any written rule — the correct response is almost always a private conversation with the member. A public correction of a community norm violation produces social embarrassment that outlasts the correction: the member who is called out for a cultural misread in front of the group often stops posting entirely, even when the original post would have been fine with minor adjustments. A private message that names the tension and explains the cultural expectation costs the operator five minutes and preserves the member.
Posting norms are the specific behavioural rules governing how content is shared: which topics belong in which channels, what requires a self-promotion disclosure, when a member needs to put content behind a link rather than posting inline, what qualifies as a direct pitch versus a helpful resource mention. Posting norm violations are often unintentional — the member posted in the wrong channel because they misread the channel structure, included a product link because they thought it was relevant, or cross-posted content without realising the community has a single-origin rule. The correct response to a posting norm violation is usually public redirection (“great topic for #channel-name”) combined with a private explanation to the member. The public redirection is logistically useful to other members who might otherwise be confused about why the post was moved or why the content disappeared; the private explanation ensures the member understands the standard and does not feel publicly corrected for an honest mistake. Posting norms benefit from being written down in a visible location — the #start-here channel, the community guide — because they are specific enough that members can check their own posts against them before posting.
Interpersonal norms govern how members treat each other in direct interactions: whether unsolicited DMs about business are permitted, how critique should be framed in public threads, whether members can follow up cold pitches with private messages, what constitutes harassment when one member’s behaviour toward another member is the issue. Interpersonal norm violations are the most delicate to handle because they involve two members with potentially conflicting accounts of what happened. The operator should never take a public position on an interpersonal dispute before speaking privately with both members. The correct initial response is a private message to both members separately, acknowledging that the operator is aware of the situation and is available to talk. Most interpersonal disputes in paid communities resolve at this stage — the act of the operator privately acknowledging the situation signals that the community takes interpersonal standards seriously, and that signal alone is often enough to change the dynamic. When the dispute does not resolve at the private-message stage, the operator needs a documented escalation path rather than an improvised response.
The practical effect of the three-tier framework is that it gives the operator a decision tree before a situation escalates. When something surfaces, the first question is which tier it belongs to: cultural, posting, or interpersonal. The answer determines the response type before the operator has to think about the specific member, the specific content, or the specific circumstances. This matters because operators who think about the specific member first tend to apply the norm inconsistently based on their relationship with that member — the most common source of the favoritism perception that damages member trust across the whole community.
3. The four-step dispute resolution ladder
When a situation requires a response, the operator needs a sequence of steps that is documented in advance and applied consistently. The four-step ladder covers the full range from minor norm violations to situations requiring permanent removal, and handles the overwhelming majority of paid community disputes without reaching the final step.
Step one: private message. Every moderation response begins here. The DM-first rule has no exceptions: no public corrections, no thread-level interventions, no redirections in the original post before a private conversation has been attempted. The private message names the specific behaviour (“the post in #general today that included the product link”), explains which community standard it conflicts with (“the posting norm is that external paid products go in #resources with a disclosure”), and describes what the operator would like to see instead. The message is written in a tone that assumes good intent. Step one resolves approximately 80% of situations in paid communities: most members who cross a line did not know exactly where the line was, and a private message from the operator that clarifies it without embarrassment is enough to permanently change the behaviour.
Step two: public redirection. Used specifically for posting norm violations where the logistical correction is useful to the community — a post in the wrong channel, content that belongs in a different format, a topic that should have a different label. Public redirection is logistical, not social: “good topic for #pricing-strategy” is a redirection; “we don’t promote paid products in this channel” is a public correction. The distinction matters because redirection helps the member and the community locate the right place for the content; public correction is primarily social and produces the embarrassment that makes members stop posting. Public redirection is always accompanied by a private message to the member explaining the reasoning behind the redirection, so that the public action and the private context arrive together rather than leaving the member to infer the reason from a public action that felt more like a correction than a logistical note.
Step three: temporary limit. Applied when a member has received two private messages about the same behaviour, in the same general category, and repeated the behaviour after both. A temporary limit restricts the member’s ability to post in specific channels for a defined period — typically seven to thirty days, depending on the nature of the behaviour. The limit is not a full mute and is not permanent. It is a formal signal that the behaviour has a documented pattern and a specific consequence, and that continuing the pattern after the limit is lifted will produce a more severe consequence. When step three is applied, the operator sends a message to the member that documents the two prior messages, names the pattern, explains the temporary limit, and states what the operator expects to see after the limit is lifted. The member now has a documented record and a clear path forward. The documentation is also the operator’s protection against later claims of arbitrary treatment.
Step four: permanent removal. Applied after step three has been completed and the behaviour has continued, or for a first-offence behaviour that is immediately harmful to a specific member. Permanent removal from a paid community is a consequential action: the member loses financial access to something they were paying for. The financial dimension means the operator should document the decision with the specific behaviour, the steps applied (or the severity rationale for immediate removal), and the refund decision. The refund question has no single correct answer, but the practice of addressing it explicitly rather than leaving it unaddressed prevents a dispute that can damage the community’s reputation when the removed member discusses their experience outside the community.
The dispute resolution ladder has a secondary function beyond moderation consistency: it changes the culture of new member onboarding. When the operator communicates the ladder to new members in the Day 0 onboarding DM or the community guide, the existence of a graduated process signals that the community takes norm enforcement seriously without being arbitrary. Members who know that a moderation response starts with a private conversation and moves through documented steps before reaching removal feel more secure posting than members who believe a single misstep could result in immediate expulsion. The paid community retention strategies guide covers the mechanisms by which member psychological safety affects contribution behaviour and, through that, month-three retention.
4. Writing enforceable community rules
Most paid community rules documents fail because they describe values rather than behaviours. “Be respectful.” “Treat others how you want to be treated.” “Add value before promoting.” These sentences describe orientations, not actions. They have no moderation basis: when a member violates them, the operator cannot quote a specific action the member took that conflicts with a specific rule — they can only express a judgment that the member’s behaviour felt disrespectful or promotional. The member who disagrees with that judgment has no documented standard to compare against. The result is a dispute about the operator’s subjective assessment rather than a documented standard consistently applied.
Enforceable rules have five properties. They are specific: the rule names a concrete action. “Do not send unsolicited DMs about your paid products or services to other members without first receiving visible consent in a public thread” is specific. “Be respectful of others’ time” is not. Specific rules can be enforced because the operator can point to the documented action (the unsolicited DM) and the documented standard (the consent requirement). They are behavioural: the rule describes something observable. “Do not follow up a cold promotional mention in a public thread with a private message to members who did not explicitly respond” describes a sequence of observable actions. “Don’t be pushy” describes an internal state that can be disagreed about. They are consistent: the same rule applies to all members regardless of tenure, community contribution, or their relationship with the operator. Applying a rule to a new member but not to a founding member who does the same thing — even once, even when the operator has a relationship-based reason for the exception — is visible to the community and produces a more corrosive effect on trust than the original violation did.
They are graduated: the rule specifies what happens on a first occurrence, a second occurrence, and an escalation. A rule that says “unsolicited product DMs will result in immediate removal” is not graduated and not credible — members who witness its enforcement will believe (and often be correct) that the operator would not actually apply immediate removal in every case. A rule that says “unsolicited product DMs will first receive a private message explaining the standard; a second occurrence will result in a seven-day posting restriction in promotional-adjacent channels; a third will result in removal” is credible because it is documented and proportionate. They are operator-modelled: the operator’s own visible behaviour in the community demonstrates the norm before it is required of members. The operator who DMs before correcting publicly, who posts with explicit context about why they are sharing something, and who gives specific feedback rather than general praise is demonstrating community standards through imitation at scale. Members who observe the operator’s posting behaviour learn more about the community’s norms from that behaviour than from any written rule document.
Three rules most paid community operators need and rarely have in documented form: the value-before-promotion rule with a specific count (“three or more non-promotional contributions visible in the community before any mention of external paid work or products” — not “add value first,” which has no enforcement basis); the DM-consent rule (“direct messages about business, collaboration, or services require visible consent in a public thread before moving to DM” — the consent visibility requirement makes the rule enforceable because the operator can see whether consent was established); and the feedback-solicitation rule (specifying whether unsolicited critique is permitted in the community or whether critique requires an explicit invitation from the member — this is the single rule whose answer most determines whether the community develops a culture of selective praise or useful feedback, and it should be explicit rather than implied). The member spotlight guide covers the contribution-first sourcing principle for spotlights; the same principle that makes a spotlight drive re-contribution (it highlights a specific observable action rather than a general endorsement) applies to the design of rules that members can understand and apply without needing the operator to interpret them.
5. Wrong moderation patterns
Five patterns appear frequently in paid community moderation and produce predictable harm that extends beyond the specific member or situation involved.
Banning on first offence for non-severe violations. Every removal from a paid community is visible in its effects: a member who disappears, a conversation that stops, an introduction post with no follow-up. Members who noticed the missing member will wonder what happened. If they learn that the removal was a first-response to a moderation issue, the information they extract is not “this community has high standards” — it is “this community removes members without warning.” The distinction matters because the referral behaviour of members who hold the second belief is significantly lower than those who hold the first. Members who believe the community enforces standards consistently and proportionately refer freely; members who believe the community removes members arbitrarily refer cautiously or not at all, because they are uncertain whether their referral will encounter the same treatment. The graduated ladder exists partly to protect the operator from this outcome.
No posted rules. Members cannot self-moderate against a standard they cannot see. When the operator’s moderation actions are the first indication most members have that a standard exists, the community has a structural norm-communication failure. Every moderation intervention in a community with no posted rules requires the operator to explain the standard as part of the intervention, which is inefficient; and the member who receives the explanation may reasonably feel that they were not given the standard before they were held to it, which produces a resentment that the operator could have avoided. A short rules document — five to eight specific behavioural rules, posted in #start-here and referenced in the Day 0 onboarding DM — eliminates the majority of first-time violations because members who can check their own posts against a clear standard do so. The paid community onboarding checklist covers the Day 0 DM format and what belongs inside the first 150 words; community standards belong in the #start-here pinned post, referenced in the DM, rather than inside the DM body where they compete for attention with the onboarding actions.
Rules written in legalese. A rules document that reads like a terms of service is a rules document that no one reads. The operators who write rules in formal legal language do so because they are imagining a worst-case scenario where the rules are disputed and need to be legally defensible. This is the wrong frame for community rules. The purpose of a community rules document is to help members understand what is expected of them so they can self-moderate. Clarity and specificity serve that purpose; legal language serves a different purpose entirely and undermines the first one. A member who has not read the rules — which is most members, when the rules are formatted as terms of service — cannot self-moderate, which means the operator ends up doing more moderation, not less.
Public call-outs. Naming a member in a public channel for a norm violation, or correcting a member’s behaviour in the thread where the violation occurred, produces an effect that outlasts the correction. Even members who agree with the operator’s assessment of the situation observe that the correction happened publicly and update their assessment of the community accordingly: this is a community where making a mistake in public produces a public consequence. That observation changes posting behaviour across the community, not just for the member who was called out. Members who were considering posting something they were uncertain about now hold back, because the cost of getting it wrong is visible. The DM-first rule exists to prevent this chilling effect. The same correction that changes a member’s behaviour when delivered privately leaves the community’s posting culture intact; the same correction delivered publicly changes both the specific member’s behaviour and the broader community’s willingness to post things they are uncertain about. The paid community member win-back guide covers the never-activated and activated-then-quiet segments; a disproportionate number of members who become permanently non-contributing after an early high-contribution period will, on close inspection, have been adjacent to a public correction event in the first month.
Inconsistent application by tenure. The moderation failure that produces the most durable damage to community trust is applying different standards to members based on their tenure, contribution level, or relationship with the operator. A founding member who self-promotes past the contribution threshold without the same consequence a new member would receive has demonstrated to every member who observed both situations that the standard is not the standard — it is the standard plus the operator’s relationship-based judgment. This is the most common source of the “favoritism” perception in small paid communities. It does not require the operator to be intentionally unfair; it requires only that the operator’s decision process for familiar members is different from their decision process for unfamiliar ones. The three-tier framework and four-step ladder reduce this risk by giving the operator a documented process to apply before their relationship-based judgment enters the picture. The Foothold community health check includes a member contribution distribution diagnostic that surfaces tenure-correlated contribution concentration, which is one of the early indicators of the conditions where inconsistent-by-tenure enforcement develops.
6. Proactive norm modelling
Most moderation problems in paid communities are preventable. The three practices that prevent the most moderation volume are all proactive rather than reactive, and all operate on the same principle: members learn community standards by observing the operator, not by reading the rules document.
The operator’s reply to a new member’s introduction post sets the tone for the entire community’s response to that member. When the operator asks a specific question that references something from the introduction, invites the member toward a specific channel or thread, and thanks them with the specificity that distinguishes acknowledgement from formality, the community observes that response and calibrates its own. New members who receive a specific, engaged response from the operator in their first post are substantially more likely to post again in the first week than those who receive a generic welcome — and the other members who observe the operator’s response pattern learn what engaged posting looks like in this community. The testimonials guide covers the member spotlight as a contribution catalysis mechanism; the operator’s reply to an introduction post is an earlier version of the same mechanism: specific acknowledgement of a specific act that makes the member feel their contribution was noticed and understood.
DM-before-correcting is a practice the operator should make visible even when it is not logistically necessary. When the operator redirects a post to the right channel publicly, then sends a private message explaining why, members who see the public redirect and later learn about the private explanation understand the sequence: public action for logistical clarity, private message for personal context. That sequence, observed even once by a member who was not involved, communicates more about the community’s moderation norms than any written rule. The operator who consistently demonstrates DM-first behaviour develops a community where members feel safe bringing concerns to the operator privately, which is the precondition for catching interpersonal disputes before they surface publicly.
The operator’s own posting behaviour is the most powerful norm-modelling mechanism available. When the operator posts with explicit context about why they are sharing something (“I’m sharing this because of a specific situation I encountered this week; what I’m looking for from this thread is responses from members who have faced the same thing”), the community learns that framing is the expected posting standard. When the operator gives specific feedback that references a concrete action (“the framing you used in paragraph two resolved something I’ve been stuck on for a month; specifically, the 'graduated response' language\rdquo;), the community learns that specific, attributed feedback is the engagement standard. These are not rules the operator has written down; they are demonstrated behaviours that members observe and replicate. The moderation framework is the structure for handling situations when those replicated behaviours fail. The proactive norm modelling is the work that determines how often the framework needs to be invoked.
Frequently asked questions
What is the difference between community norms and community rules for a paid community?
Community norms are the cultural expectations that govern how the community feels and behaves: what kinds of sharing are valued, how critique is given, what the community’s relationship to self-promotion is. Community rules are the specific, documented behavioural standards that enforce those norms. Norms exist whether or not they are written down — every paid community has a culture, and new members infer the norms from watching how the operator and tenured members behave. Rules exist only when they are written, communicated, and consistently applied. The failure mode for most operators is having strong norms but no rules: members feel the expectations without having a documented standard to consult, so every moderation action feels arbitrary even to members who agree with the decision. The three-tier framework separates these into community norms (cultural values requiring private conversation when violated), posting norms (specific behavioural rules requiring public redirection plus private explanation), and interpersonal norms (member-to-member interaction standards requiring private conversation with both parties).
How do you handle a member dispute in a paid community?
The four-step resolution ladder handles the overwhelming majority of paid community disputes. Step one: private message to the member naming the specific behaviour, the community standard it conflicts with, and what the operator would like to see instead. This resolves about 80% of situations because most members who cross a line did not know where the line was. Step two: public redirection (logistical correction to the right channel or format) combined with a private explanation, used specifically for posting norm violations where the public correction helps the broader community understand context. Step three: temporary limit restricting posting in specific channels after a member has received two private messages about the same behaviour and repeated it. Step four: permanent removal, applied after step three has been exhausted or for first-offence behaviour that is immediately harmful to another member. Because paid community membership involves financial access, every step-three and step-four action should be documented in a moderation log with the date, the specific behaviour, the steps applied, and the refund decision.
What should paid community rules include?
Enforceable rules have five properties: specific (names a concrete action, not a value), behavioural (describes something observable), consistent (same rule applies to all members regardless of tenure), graduated (specifies what happens on first, second, and third occurrences), and operator-modelled (the operator’s own posting behaviour demonstrates the norm). The three rules most paid communities need and rarely have in documented form: (1) the value-before-promotion rule with a specific count (three non-promotional contributions before any mention of external paid work — “add value first” has no enforcement basis); (2) the DM-consent rule (direct messages about business or services require visible consent in a public thread before the member moves to DM — the consent visibility makes the rule enforceable); (3) the feedback-solicitation rule specifying whether unsolicited critique is permitted or requires an explicit invitation (the answer to this question most determines whether the community develops a culture of useful feedback or selective praise). Rules written in legal language, or framed as values rather than actions, are not read and cannot be enforced consistently.
When should you ban a member from a paid community?
Permanent removal is appropriate in two situations: after the three preceding steps in the dispute resolution ladder have been applied and the behaviour has continued, or for a first-offence behaviour that is immediately harmful to a specific member. Because paid community membership involves financial access, permanent removal should always be documented: the specific behaviour, the steps applied, and the refund decision. The refund question has no single correct answer, but addressing it explicitly prevents a dispute that can damage the community’s reputation when the removed member discusses their experience publicly. Banning on first offence for non-severe violations — even when the behaviour was clearly wrong — produces a specific harm: members who learn about the first-offence removal update their assessment of the community from “high standards, proportionate enforcement” to “arbitrary removal” and reduce their referral behaviour accordingly. The graduated ladder exists partly to prevent this outcome.
How do you prevent moderation problems before they happen in a paid community?
Three proactive practices prevent most moderation volume. First, post specific behavioural rules where members can see them before they post: #start-here pinned post, referenced in the Day 0 onboarding DM. Members who can check their own posts against clear, specific rules self-moderate; members who discover the standard for the first time when they’ve violated it feel set up to fail. Second, model the norms through the operator’s own visible posting behaviour: the specificity of the operator’s replies to new member introductions, the way the operator frames feedback, the operator’s DM-before-correcting practice — all teach community standards through imitation more effectively than any written rule. Third, engage actively with new members in the first week: most moderation problems that reach step three or four started as early posting-norm mismatches that were never addressed and became entrenched. A private message in day three or day seven, when the pattern first appears, costs five minutes and prevents the situation that costs five hours at month three.