{"content_id":"cyaoee3j4i","slug":"creator-community-nine-month-operation-case-study","locale":"en","schema_type":"Article","category":"case_study","category_name":"Case Study","title":"A 9-Month Creator Community Case Study and Key Lessons","summary":"Drawing on roughly 87,000 messages accumulated while design creator Iki ran a community for about nine months, this article outlines its growth stages, the problem of dependence on the operator, the role of core members, and the conditions for self-organized groups. The key lesson from the case is that relationships among members and repeat participation matter more than the number of people who join.","author":{"name":"Injoys Editorial Team","url":"https://injoys.com/ko/about"},"key_points":["Follower count indicates the reach of content, but it does not directly prove the depth of relationships within a community or members' willingness to participate.","A community in which the operator starts most conversations may appear active, but it can easily stagnate if the operator stops participating.","In this case, the three most active core members played an important role in welcoming new members, generating questions, and sustaining conversations.","A modest onboarding process and clear participation criteria can be more useful than simply increasing headcount when seeking members with a strong willingness to participate.","When a community begins organizing its own study groups and gatherings, it can be seen as a sign that it is evolving from an operator-centered channel into a member-centered network."],"content_markdown":"Even as content views and follower counts grow, the relationship between creators and users does not necessarily deepen. The nine-month operating case of Iki, a solo creator who shares design records, shows what should be measured and changed when converting content viewers into members who interact repeatedly.\n\nThe figures for people, messages, and conversions in this article are case data compiled from records disclosed by the operator. They are not independently verified experimental results or standards that apply equally to every community.\n\n## Background of the Case and Reasons for Starting the Community\n\nIki is a solo creator who produces design-related content on Instagram and YouTube and has used vibe coding to build Toksignal, a website that analyzes signals in KakaoTalk conversations, and Munchae, an app for saving sentences. When the community was launched, the Instagram follower count was described as approximately 14,000–15,000.\n\nEven as the number of followers increased, only some people left DMs and comments. This experience revealed two issues.\n\n1. **The difference between reach and relationships:** A follow or view signals exposure to content, but it does not indicate an ongoing conversation or collaborative relationship.\n2. **Platform dependence:** If algorithms, exposure policies, or account status change, the reach accumulated on a specific platform may be affected.\n\nA more direct motivation for creating the community was the isolation of developing ideas alone, publishing them, and then waiting for a response. The goal was not simply to increase subscriber numbers on another channel, but to share the growth process with people who had similar interests.\n\n## Key Data from the Nine-Month Operating Case\n\n| Item | Details reported in the case | Points to consider when interpreting |\n|---|---:|---|\n| Community operating period | Approximately 9 months | Exact start and end dates were not provided |\n| Cumulative messages | Approximately 87,000 | Message count does not directly indicate conversation quality or participant diversity |\n| Participants in the first week | Approximately 10 | Primarily previous challenge participants and acquaintances |\n| Mid-stage size | Approximately 50 | Size reached after continued content-based promotion |\n| Short-term influx | Approximately 30 over 2 days | Increased from 50 to approximately 80 after a specific Reels post |\n| Estimated conversion rate | Approximately 0.21% | A simple calculation dividing 30 by 14,000 followers, not a conversion rate based on exposed users |\n| Core active members | Top 3 | Observed to play a major role in welcoming new members and facilitating conversations |\n\nThe 0.21% figure is a case-specific calculation that uses the total follower count as the denominator. To assess actual conversion performance, the number of people exposed to the post, those who clicked the link, those who completed an application, and those who remained after a certain period must each be measured.\n\n## How Followers and Community Members Differ\n\nFollowers primarily discover and consume content. By contrast, becoming a community member requires turning interest into action.\n\nA typical participation journey is as follows.\n\n1. Discover the content.\n2. Become interested in the topic or creator.\n3. Open the profile or information link.\n4. Review the community’s purpose and conditions.\n5. Join or complete an application.\n6. Take a first action, such as introducing oneself, asking a question, or answering.\n7. Return and build relationships with other members.\n\nBecause people drop off at every stage, growth in followers and growth in community membership may not be proportional. In particular, people who joined but took no first action must be distinguished from those who participate repeatedly.\n\n| Category | Follower-centered channel | Relationship-centered community |\n|---|---|---|\n| Primary actions | Viewing, liking, sharing | Asking questions, answering, collaborating, creating gatherings |\n| Relationship structure | Creator to many | Member to member |\n| Representative metrics | Reach, views, follower growth | Active participants, return visits, responses, member-led activities |\n| Value producers | Primarily the creator | Both the operator and members |\n| Primary risks | Dependence on platforms and algorithms | Dependence on the operator, conflict, core-member burnout |\n\n## Growth and Stagnation During the First Six Months\n\n### Months 1–3: Activation Stage\n\nInitially, new sign-ups, introductions, questions, and answers increased. The anticipation and desire to explore a new space can be interpreted as having facilitated conversation. Early interactions also developed as existing members actively answered questions from new members.\n\n### Months 4–6: Declining Novelty and Repetition\n\nOver time, the influx of new members slowed and conversation topics became repetitive. Some members only read messages, and the overall number of messages also declined. This does not necessarily indicate failure, but it is a signal to review the following items.\n\n- Time required for a new member to take a first action\n- Reparticipation rates 7, 30, and 90 days after joining\n- Proportion of conversations among members that exclude the operator\n- Time from posting a question to receiving the first answer\n- Whether only the same small group is speaking\n- Whether the community’s purpose aligns with its current conversation topics\n\nMembers who only read may still gain information or participate later, so inactivity should not immediately be judged as having no value. Uneven participation is common in online communities, and the so-called 90-9-1 rule is not an exact law but closer to an empirical heuristic describing this tendency.\n\n## Structural Problems Created by Operator-Centered Conversations\n\nEarly conversations generally followed this sequence.\n\n\u003e Operator’s question → Member’s answer → Operator introduces a new topic → Operator’s follow-up response\n\nThe community may appear active based on message count alone, but in this structure, the conversation also stops when the operator stops. The operator in this case assessed it as being closer to a “one-person show” than a community.\n\nTo assess dependence on the operator, the following ratios are more useful than the total number of messages.\n\n- **Operator message share:** Number of operator messages ÷ total number of messages\n- **Member-to-member response ratio:** Number of answers exchanged among members without operator involvement ÷ total number of answers\n- **Member-initiated conversation ratio:** Number of conversation threads initiated by members ÷ total number of conversation threads\n- **Unanswered question ratio:** Number of questions that received no answer ÷ total number of questions\n\nThe right direction is not for the operator to disappear, but to change roles. Rather than initiating every conversation, the operator should become more like a rule designer, connector, conflict mediator, and guide who helps new members adapt initially.\n\n## How Core Members Kept the Community Moving\n\nAn analysis of nine months of messages reportedly found that the top three most active members performed the following roles.\n\n- Greeting new members first\n- Responding to unanswered questions\n- Suggesting new topics\n- Asking follow-up questions to keep conversations going\n- Naturally communicating the community’s atmosphere and practices\n\nThey play a role similar to a service’s power users. However, people who send many messages should not immediately be considered healthy core members. It is also necessary to assess whether they help others participate, avoid monopolizing conversations, and respect the rules.\n\nCore members may burn out if they remain responsible for welcoming others and answering questions. Mechanisms are needed to distribute the burden and recognize contributions, such as rotating welcome duties, granting authority to lead study groups, offering the option to take breaks, and giving public thanks.\n\n## Changes to the Sign-Up Process and Inactive-Member Management\n\n### Switching from Immediate Entry to a Simple Application Process\n\nInitially, anyone could enter the KakaoTalk open chat room simply by clicking a link. Access was easy, but this also resulted in members with low interest and rapid departures. A later application process containing a minimal set of questions was introduced to confirm each person’s purpose for participating.\n\nRather than making sign-up questions lengthy, it is better to request only information that will actually be used in operations.\n\n- What topics are you interested in?\n- What do you hope to gain from the community?\n- What experiences can you share with other members?\n- Do you agree to the basic rules and privacy notice?\n\nNeither the principle that lower entry barriers are always better nor the principle that higher barriers always produce better members is absolute. The appropriate level should be identified by comparing application completion rates with post-enrollment participation rates.\n\n### Removing Inactive Members\n\nThe operator gave advance notice to members who had been inactive for an extended period and removed accounts that did not respond. It was reported that this increased the activity rate of the remaining members and the community’s density, but the extent of the improvement cannot be confirmed because before-and-after figures were not disclosed.\n\nIt is safer to remove inactive members only after clarifying the following conditions.\n\n1. Disclose the definition of activity and the review period in advance.\n2. Explain whether read-only participation is permitted.\n3. Provide an opportunity for members to indicate their wishes before removal.\n4. Consider temporary personal circumstances involving health, work, caregiving, and other matters.\n5. Explain the process for rejoining after departure.\n\nThe goal is not to reduce the numbers arbitrarily, but to create a participation structure suited to the community’s current purpose.\n\n## Signals of a Transition to a Self-Sustaining Community\n\nAn important turning point was the creation of English, lettering, Japanese, and marketing study groups proposed by members rather than planned by the operator. The operator supported the space and recruitment, while members handled the actual planning and facilitation.\n\nThe following phenomena may signal growing self-sustainability.\n\n- Members initiate questions and propose events.\n- Questions receive answers even without the operator.\n- Members exchange knowledge, feedback, and opportunities with one another.\n- Small-group activities recur rather than ending after a single session.\n- Existing members help new members adapt.\n- Problems are resolved according to shared rules.\n\nSelf-sustainability is not the same as having no management. The operator must maintain safety rules, reporting procedures, privacy protection, and standards governing advertising and conflicts of interest.\n\n## Validating MVP Demand Through the Community\n\nIn this case, the community was also used as an early user-research space to assess the idea for a paid challenge. The idea was shared, specific feedback was collected, and the program was revised before launch.\n\nThe advantage of community feedback is that it is rich in context and allows follow-up questions. However, it also has the following biases.\n\n- Many participants already have a favorable view of the creator.\n- The vocal minority may not represent overall demand.\n- Interest in a free offering may differ from actual willingness to pay.\n- Negative opinions may be less common in public conversations.\n\nAccordingly, MVP validation should progressively confirm actions rather than statements.\n\n| Validation stage | Signal to confirm |\n|---|---|\n| Problem validation | Does the same problem recur among multiple people? |\n| Solution feedback | Are specific suggestions for revision offered? |\n| Willingness to participate | Do people ask about the schedule, conditions, and required materials? |\n| Willingness to act | Do people actually sign up for a waitlist or test? |\n| Willingness to pay | Does willingness to purchase or reserve remain after the price is disclosed? |\n| Ongoing value | Do repeat use, recommendations, or follow-up activities occur after participation? |\n\nTo reduce selection bias, research should not be limited to community members. It should also include inactive members, people who left, and prospective customers outside the community.\n\n## Key Metrics for Community Operators to Track\n\nIt is difficult to assess community health using only member and message counts. Depending on the purpose, the following metrics can also be recorded.\n\n| Metric | Calculation or observation method | Meaning |\n|---|---|---|\n| Sign-up conversion rate | Completed sign-ups ÷ visitors to the information link | Suitability of the sign-up process |\n| First-action rate | New members who took a first action ÷ new sign-ups | Onboarding effectiveness |\n| Period activity rate | Members who acted during the period ÷ all eligible members | Current breadth of participation |\n| Reparticipation rate | Members who participated during at least two periods ÷ participants in the previous period | Level of repeated participation |\n| Member-to-member response rate | Questions answered by members ÷ all answered questions | Independence from the operator |\n| Response time | Median time from a question to the first valid answer | Speed of receiving help |\n| Participation concentration | Share of messages from the most active participants | Risk of dependence on a small group |\n| Number of member-led activities | Number of activities proposed and operated by members | Self-sustainability |\n| Reasons for leaving | Categories identified through departure surveys or interviews | Clues for structural improvement |\n\nEven when message volume is high, satisfaction may be low if there is extensive small talk, duplicate notifications, or domination by the operator. Quantitative metrics should be supplemented with short interviews, anonymous surveys, and reviews of reasons for leaving.\n\n## Privacy Principles for Analyzing Message Data\n\nCommunity conversations may contain information about participants’ relationships, interests, schedules, or personal concerns. The following principles are necessary when using them for operational analysis.\n\n- Inform members of the purpose of collection and scope of analysis.\n- Remove names, contact details, and account identifiers that are unnecessary for the analysis.\n- Do not use private messages as public examples without consent.\n- Limit access to original messages and set a retention period.\n- Before entering conversations into external AI tools, review the relevant service’s data-processing terms.\n- Present public findings in aggregate units that do not allow individuals to be identified.\n\nThe platform’s terms of service and applicable privacy regulations must also be reviewed separately. The convenience of community analysis cannot take precedence over members’ privacy and trust.\n\n## Operating Principles Derived from the Nine-Month Case\n\n### Start Small\n\nA community can be tested without a large follower base. Rather than gathering hundreds of people from the outset, securing 10 people who repeatedly converse and help one another makes it easier to learn how the operating structure works.\n\n### Assume Iterative Improvement Rather Than a Finished Design\n\nSign-up conditions, channel structure, roles, and inactivity criteria should be adjusted based on actual behavioral data. Explaining the reason for a change and its implementation date, then reviewing the results after a set period, can reduce arbitrary management.\n\n### The Operator Is a Designer of Participation Structures, Not a Producer of Conversations\n\nInstead of answering every question first, the operator should give members time to respond. It is necessary to connect people with similar interests and give members the authority to execute activities they propose.\n\n### Examine the Distribution of Relationships Rather Than the Total Number of People\n\nIn addition to member count, assess how many people initiate conversations, whether different people provide answers, and whether core members are becoming exhausted. A healthy community has multiple connections rather than a single star operator.\n\n## A 30-Day Review Plan for Implementation\n\n- **Week 1:** Define the community’s target audience, value provided, and permitted and prohibited behaviors in one sentence each.\n- **Week 2:** Document the journey from joining to the first introduction, first answer, and return visit, then identify drop-off points.\n- **Week 3:** Identify core members who help others and discuss their roles and burdens.\n- **Week 4:** Compare the operator’s message share, member-to-member responses, and member-led activities.\n- **End of the month:** Decide which experiments to continue and which to stop, then explain the changes to members.\n\nThe key lesson from this case is that “gathering many people” and “helping people build relationships with one another” are different tasks. Community growth should be judged not only by a graph of member numbers, but also by whether questions, answers, and collaboration continue even when the operator temporarily steps back.","content_html":"\u003cp\u003eEven as content views and follower counts grow, the relationship between creators and users does not necessarily deepen. The nine-month operating case of Iki, a solo creator who shares design records, shows what should be measured and changed when converting content viewers into members who interact repeatedly.\u003c/p\u003e\n\u003cp\u003eThe figures for people, messages, and conversions in this article are case data compiled from records disclosed by the operator. They are not independently verified experimental results or standards that apply equally to every community.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#background-of-the-case-and-reasons-for-starting-the-community\" class=\"anchor\" id=\"background-of-the-case-and-reasons-for-starting-the-community\"\u003e\u003c/a\u003eBackground of the Case and Reasons for Starting the Community\u003c/h2\u003e\n\u003cp\u003eIki is a solo creator who produces design-related content on Instagram and YouTube and has used vibe coding to build Toksignal, a website that analyzes signals in KakaoTalk conversations, and Munchae, an app for saving sentences. When the community was launched, the Instagram follower count was described as approximately 14,000–15,000.\u003c/p\u003e\n\u003cp\u003eEven as the number of followers increased, only some people left DMs and comments. This experience revealed two issues.\u003c/p\u003e\n\u003col\u003e\n\u003cli\u003e\n\u003cstrong\u003eThe difference between reach and relationships:\u003c/strong\u003e A follow or view signals exposure to content, but it does not indicate an ongoing conversation or collaborative relationship.\u003c/li\u003e\n\u003cli\u003e\n\u003cstrong\u003ePlatform dependence:\u003c/strong\u003e If algorithms, exposure policies, or account status change, the reach accumulated on a specific platform may be affected.\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eA more direct motivation for creating the community was the isolation of developing ideas alone, publishing them, and then waiting for a response. The goal was not simply to increase subscriber numbers on another channel, but to share the growth process with people who had similar interests.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#key-data-from-the-nine-month-operating-case\" class=\"anchor\" id=\"key-data-from-the-nine-month-operating-case\"\u003e\u003c/a\u003eKey Data from the Nine-Month Operating Case\u003c/h2\u003e\n\u003cdiv class=\"overflow-x-auto\"\u003e\u003ctable\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth\u003eItem\u003c/th\u003e\n\u003cth\u003eDetails reported in the case\u003c/th\u003e\n\u003cth\u003ePoints to consider when interpreting\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Item\"\u003eCommunity operating period\u003c/td\u003e\n\u003ctd data-label=\"Details reported in the case\"\u003eApproximately 9 months\u003c/td\u003e\n\u003ctd data-label=\"Points to consider when interpreting\"\u003eExact start and end dates were not provided\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Item\"\u003eCumulative messages\u003c/td\u003e\n\u003ctd data-label=\"Details reported in the case\"\u003eApproximately 87,000\u003c/td\u003e\n\u003ctd data-label=\"Points to consider when interpreting\"\u003eMessage count does not directly indicate conversation quality or participant diversity\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Item\"\u003eParticipants in the first week\u003c/td\u003e\n\u003ctd data-label=\"Details reported in the case\"\u003eApproximately 10\u003c/td\u003e\n\u003ctd data-label=\"Points to consider when interpreting\"\u003ePrimarily previous challenge participants and acquaintances\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Item\"\u003eMid-stage size\u003c/td\u003e\n\u003ctd data-label=\"Details reported in the case\"\u003eApproximately 50\u003c/td\u003e\n\u003ctd data-label=\"Points to consider when interpreting\"\u003eSize reached after continued content-based promotion\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Item\"\u003eShort-term influx\u003c/td\u003e\n\u003ctd data-label=\"Details reported in the case\"\u003eApproximately 30 over 2 days\u003c/td\u003e\n\u003ctd data-label=\"Points to consider when interpreting\"\u003eIncreased from 50 to approximately 80 after a specific Reels post\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Item\"\u003eEstimated conversion rate\u003c/td\u003e\n\u003ctd data-label=\"Details reported in the case\"\u003eApproximately 0.21%\u003c/td\u003e\n\u003ctd data-label=\"Points to consider when interpreting\"\u003eA simple calculation dividing 30 by 14,000 followers, not a conversion rate based on exposed users\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Item\"\u003eCore active members\u003c/td\u003e\n\u003ctd data-label=\"Details reported in the case\"\u003eTop 3\u003c/td\u003e\n\u003ctd data-label=\"Points to consider when interpreting\"\u003eObserved to play a major role in welcoming new members and facilitating conversations\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\u003c/div\u003e\n\u003cp\u003eThe 0.21% figure is a case-specific calculation that uses the total follower count as the denominator. To assess actual conversion performance, the number of people exposed to the post, those who clicked the link, those who completed an application, and those who remained after a certain period must each be measured.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#how-followers-and-community-members-differ\" class=\"anchor\" id=\"how-followers-and-community-members-differ\"\u003e\u003c/a\u003eHow Followers and Community Members Differ\u003c/h2\u003e\n\u003cp\u003eFollowers primarily discover and consume content. By contrast, becoming a community member requires turning interest into action.\u003c/p\u003e\n\u003cp\u003eA typical participation journey is as follows.\u003c/p\u003e\n\u003col\u003e\n\u003cli\u003eDiscover the content.\u003c/li\u003e\n\u003cli\u003eBecome interested in the topic or creator.\u003c/li\u003e\n\u003cli\u003eOpen the profile or information link.\u003c/li\u003e\n\u003cli\u003eReview the community’s purpose and conditions.\u003c/li\u003e\n\u003cli\u003eJoin or complete an application.\u003c/li\u003e\n\u003cli\u003eTake a first action, such as introducing oneself, asking a question, or answering.\u003c/li\u003e\n\u003cli\u003eReturn and build relationships with other members.\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eBecause people drop off at every stage, growth in followers and growth in community membership may not be proportional. In particular, people who joined but took no first action must be distinguished from those who participate repeatedly.\u003c/p\u003e\n\u003cdiv class=\"overflow-x-auto\"\u003e\u003ctable\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth\u003eCategory\u003c/th\u003e\n\u003cth\u003eFollower-centered channel\u003c/th\u003e\n\u003cth\u003eRelationship-centered community\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Category\"\u003ePrimary actions\u003c/td\u003e\n\u003ctd data-label=\"Follower-centered channel\"\u003eViewing, liking, sharing\u003c/td\u003e\n\u003ctd data-label=\"Relationship-centered community\"\u003eAsking questions, answering, collaborating, creating gatherings\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Category\"\u003eRelationship structure\u003c/td\u003e\n\u003ctd data-label=\"Follower-centered channel\"\u003eCreator to many\u003c/td\u003e\n\u003ctd data-label=\"Relationship-centered community\"\u003eMember to member\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Category\"\u003eRepresentative metrics\u003c/td\u003e\n\u003ctd data-label=\"Follower-centered channel\"\u003eReach, views, follower growth\u003c/td\u003e\n\u003ctd data-label=\"Relationship-centered community\"\u003eActive participants, return visits, responses, member-led activities\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Category\"\u003eValue producers\u003c/td\u003e\n\u003ctd data-label=\"Follower-centered channel\"\u003ePrimarily the creator\u003c/td\u003e\n\u003ctd data-label=\"Relationship-centered community\"\u003eBoth the operator and members\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Category\"\u003ePrimary risks\u003c/td\u003e\n\u003ctd data-label=\"Follower-centered channel\"\u003eDependence on platforms and algorithms\u003c/td\u003e\n\u003ctd data-label=\"Relationship-centered community\"\u003eDependence on the operator, conflict, core-member burnout\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\u003c/div\u003e\n\u003ch2\u003e\n\u003ca href=\"#growth-and-stagnation-during-the-first-six-months\" class=\"anchor\" id=\"growth-and-stagnation-during-the-first-six-months\"\u003e\u003c/a\u003eGrowth and Stagnation During the First Six Months\u003c/h2\u003e\n\u003ch3\u003e\n\u003ca href=\"#months-13-activation-stage\" class=\"anchor\" id=\"months-13-activation-stage\"\u003e\u003c/a\u003eMonths 1–3: Activation Stage\u003c/h3\u003e\n\u003cp\u003eInitially, new sign-ups, introductions, questions, and answers increased. The anticipation and desire to explore a new space can be interpreted as having facilitated conversation. Early interactions also developed as existing members actively answered questions from new members.\u003c/p\u003e\n\u003ch3\u003e\n\u003ca href=\"#months-46-declining-novelty-and-repetition\" class=\"anchor\" id=\"months-46-declining-novelty-and-repetition\"\u003e\u003c/a\u003eMonths 4–6: Declining Novelty and Repetition\u003c/h3\u003e\n\u003cp\u003eOver time, the influx of new members slowed and conversation topics became repetitive. Some members only read messages, and the overall number of messages also declined. This does not necessarily indicate failure, but it is a signal to review the following items.\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eTime required for a new member to take a first action\u003c/li\u003e\n\u003cli\u003eReparticipation rates 7, 30, and 90 days after joining\u003c/li\u003e\n\u003cli\u003eProportion of conversations among members that exclude the operator\u003c/li\u003e\n\u003cli\u003eTime from posting a question to receiving the first answer\u003c/li\u003e\n\u003cli\u003eWhether only the same small group is speaking\u003c/li\u003e\n\u003cli\u003eWhether the community’s purpose aligns with its current conversation topics\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eMembers who only read may still gain information or participate later, so inactivity should not immediately be judged as having no value. Uneven participation is common in online communities, and the so-called 90-9-1 rule is not an exact law but closer to an empirical heuristic describing this tendency.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#structural-problems-created-by-operator-centered-conversations\" class=\"anchor\" id=\"structural-problems-created-by-operator-centered-conversations\"\u003e\u003c/a\u003eStructural Problems Created by Operator-Centered Conversations\u003c/h2\u003e\n\u003cp\u003eEarly conversations generally followed this sequence.\u003c/p\u003e\n\u003cblockquote\u003e\n\u003cp\u003eOperator’s question → Member’s answer → Operator introduces a new topic → Operator’s follow-up response\u003c/p\u003e\n\u003c/blockquote\u003e\n\u003cp\u003eThe community may appear active based on message count alone, but in this structure, the conversation also stops when the operator stops. The operator in this case assessed it as being closer to a “one-person show” than a community.\u003c/p\u003e\n\u003cp\u003eTo assess dependence on the operator, the following ratios are more useful than the total number of messages.\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003e\n\u003cstrong\u003eOperator message share:\u003c/strong\u003e Number of operator messages ÷ total number of messages\u003c/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eMember-to-member response ratio:\u003c/strong\u003e Number of answers exchanged among members without operator involvement ÷ total number of answers\u003c/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eMember-initiated conversation ratio:\u003c/strong\u003e Number of conversation threads initiated by members ÷ total number of conversation threads\u003c/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eUnanswered question ratio:\u003c/strong\u003e Number of questions that received no answer ÷ total number of questions\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eThe right direction is not for the operator to disappear, but to change roles. Rather than initiating every conversation, the operator should become more like a rule designer, connector, conflict mediator, and guide who helps new members adapt initially.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#how-core-members-kept-the-community-moving\" class=\"anchor\" id=\"how-core-members-kept-the-community-moving\"\u003e\u003c/a\u003eHow Core Members Kept the Community Moving\u003c/h2\u003e\n\u003cp\u003eAn analysis of nine months of messages reportedly found that the top three most active members performed the following roles.\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eGreeting new members first\u003c/li\u003e\n\u003cli\u003eResponding to unanswered questions\u003c/li\u003e\n\u003cli\u003eSuggesting new topics\u003c/li\u003e\n\u003cli\u003eAsking follow-up questions to keep conversations going\u003c/li\u003e\n\u003cli\u003eNaturally communicating the community’s atmosphere and practices\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eThey play a role similar to a service’s power users. However, people who send many messages should not immediately be considered healthy core members. It is also necessary to assess whether they help others participate, avoid monopolizing conversations, and respect the rules.\u003c/p\u003e\n\u003cp\u003eCore members may burn out if they remain responsible for welcoming others and answering questions. Mechanisms are needed to distribute the burden and recognize contributions, such as rotating welcome duties, granting authority to lead study groups, offering the option to take breaks, and giving public thanks.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#changes-to-the-sign-up-process-and-inactive-member-management\" class=\"anchor\" id=\"changes-to-the-sign-up-process-and-inactive-member-management\"\u003e\u003c/a\u003eChanges to the Sign-Up Process and Inactive-Member Management\u003c/h2\u003e\n\u003ch3\u003e\n\u003ca href=\"#switching-from-immediate-entry-to-a-simple-application-process\" class=\"anchor\" id=\"switching-from-immediate-entry-to-a-simple-application-process\"\u003e\u003c/a\u003eSwitching from Immediate Entry to a Simple Application Process\u003c/h3\u003e\n\u003cp\u003eInitially, anyone could enter the KakaoTalk open chat room simply by clicking a link. Access was easy, but this also resulted in members with low interest and rapid departures. A later application process containing a minimal set of questions was introduced to confirm each person’s purpose for participating.\u003c/p\u003e\n\u003cp\u003eRather than making sign-up questions lengthy, it is better to request only information that will actually be used in operations.\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eWhat topics are you interested in?\u003c/li\u003e\n\u003cli\u003eWhat do you hope to gain from the community?\u003c/li\u003e\n\u003cli\u003eWhat experiences can you share with other members?\u003c/li\u003e\n\u003cli\u003eDo you agree to the basic rules and privacy notice?\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eNeither the principle that lower entry barriers are always better nor the principle that higher barriers always produce better members is absolute. The appropriate level should be identified by comparing application completion rates with post-enrollment participation rates.\u003c/p\u003e\n\u003ch3\u003e\n\u003ca href=\"#removing-inactive-members\" class=\"anchor\" id=\"removing-inactive-members\"\u003e\u003c/a\u003eRemoving Inactive Members\u003c/h3\u003e\n\u003cp\u003eThe operator gave advance notice to members who had been inactive for an extended period and removed accounts that did not respond. It was reported that this increased the activity rate of the remaining members and the community’s density, but the extent of the improvement cannot be confirmed because before-and-after figures were not disclosed.\u003c/p\u003e\n\u003cp\u003eIt is safer to remove inactive members only after clarifying the following conditions.\u003c/p\u003e\n\u003col\u003e\n\u003cli\u003eDisclose the definition of activity and the review period in advance.\u003c/li\u003e\n\u003cli\u003eExplain whether read-only participation is permitted.\u003c/li\u003e\n\u003cli\u003eProvide an opportunity for members to indicate their wishes before removal.\u003c/li\u003e\n\u003cli\u003eConsider temporary personal circumstances involving health, work, caregiving, and other matters.\u003c/li\u003e\n\u003cli\u003eExplain the process for rejoining after departure.\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eThe goal is not to reduce the numbers arbitrarily, but to create a participation structure suited to the community’s current purpose.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#signals-of-a-transition-to-a-self-sustaining-community\" class=\"anchor\" id=\"signals-of-a-transition-to-a-self-sustaining-community\"\u003e\u003c/a\u003eSignals of a Transition to a Self-Sustaining Community\u003c/h2\u003e\n\u003cp\u003eAn important turning point was the creation of English, lettering, Japanese, and marketing study groups proposed by members rather than planned by the operator. The operator supported the space and recruitment, while members handled the actual planning and facilitation.\u003c/p\u003e\n\u003cp\u003eThe following phenomena may signal growing self-sustainability.\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eMembers initiate questions and propose events.\u003c/li\u003e\n\u003cli\u003eQuestions receive answers even without the operator.\u003c/li\u003e\n\u003cli\u003eMembers exchange knowledge, feedback, and opportunities with one another.\u003c/li\u003e\n\u003cli\u003eSmall-group activities recur rather than ending after a single session.\u003c/li\u003e\n\u003cli\u003eExisting members help new members adapt.\u003c/li\u003e\n\u003cli\u003eProblems are resolved according to shared rules.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eSelf-sustainability is not the same as having no management. The operator must maintain safety rules, reporting procedures, privacy protection, and standards governing advertising and conflicts of interest.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#validating-mvp-demand-through-the-community\" class=\"anchor\" id=\"validating-mvp-demand-through-the-community\"\u003e\u003c/a\u003eValidating MVP Demand Through the Community\u003c/h2\u003e\n\u003cp\u003eIn this case, the community was also used as an early user-research space to assess the idea for a paid challenge. The idea was shared, specific feedback was collected, and the program was revised before launch.\u003c/p\u003e\n\u003cp\u003eThe advantage of community feedback is that it is rich in context and allows follow-up questions. However, it also has the following biases.\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eMany participants already have a favorable view of the creator.\u003c/li\u003e\n\u003cli\u003eThe vocal minority may not represent overall demand.\u003c/li\u003e\n\u003cli\u003eInterest in a free offering may differ from actual willingness to pay.\u003c/li\u003e\n\u003cli\u003eNegative opinions may be less common in public conversations.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eAccordingly, MVP validation should progressively confirm actions rather than statements.\u003c/p\u003e\n\u003cdiv class=\"overflow-x-auto\"\u003e\u003ctable\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth\u003eValidation stage\u003c/th\u003e\n\u003cth\u003eSignal to confirm\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Validation stage\"\u003eProblem validation\u003c/td\u003e\n\u003ctd data-label=\"Signal to confirm\"\u003eDoes the same problem recur among multiple people?\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Validation stage\"\u003eSolution feedback\u003c/td\u003e\n\u003ctd data-label=\"Signal to confirm\"\u003eAre specific suggestions for revision offered?\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Validation stage\"\u003eWillingness to participate\u003c/td\u003e\n\u003ctd data-label=\"Signal to confirm\"\u003eDo people ask about the schedule, conditions, and required materials?\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Validation stage\"\u003eWillingness to act\u003c/td\u003e\n\u003ctd data-label=\"Signal to confirm\"\u003eDo people actually sign up for a waitlist or test?\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Validation stage\"\u003eWillingness to pay\u003c/td\u003e\n\u003ctd data-label=\"Signal to confirm\"\u003eDoes willingness to purchase or reserve remain after the price is disclosed?\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Validation stage\"\u003eOngoing value\u003c/td\u003e\n\u003ctd data-label=\"Signal to confirm\"\u003eDo repeat use, recommendations, or follow-up activities occur after participation?\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\u003c/div\u003e\n\u003cp\u003eTo reduce selection bias, research should not be limited to community members. It should also include inactive members, people who left, and prospective customers outside the community.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#key-metrics-for-community-operators-to-track\" class=\"anchor\" id=\"key-metrics-for-community-operators-to-track\"\u003e\u003c/a\u003eKey Metrics for Community Operators to Track\u003c/h2\u003e\n\u003cp\u003eIt is difficult to assess community health using only member and message counts. Depending on the purpose, the following metrics can also be recorded.\u003c/p\u003e\n\u003cdiv class=\"overflow-x-auto\"\u003e\u003ctable\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth\u003eMetric\u003c/th\u003e\n\u003cth\u003eCalculation or observation method\u003c/th\u003e\n\u003cth\u003eMeaning\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Metric\"\u003eSign-up conversion rate\u003c/td\u003e\n\u003ctd data-label=\"Calculation or observation method\"\u003eCompleted sign-ups ÷ visitors to the information link\u003c/td\u003e\n\u003ctd data-label=\"Meaning\"\u003eSuitability of the sign-up process\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Metric\"\u003eFirst-action rate\u003c/td\u003e\n\u003ctd data-label=\"Calculation or observation method\"\u003eNew members who took a first action ÷ new sign-ups\u003c/td\u003e\n\u003ctd data-label=\"Meaning\"\u003eOnboarding effectiveness\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Metric\"\u003ePeriod activity rate\u003c/td\u003e\n\u003ctd data-label=\"Calculation or observation method\"\u003eMembers who acted during the period ÷ all eligible members\u003c/td\u003e\n\u003ctd data-label=\"Meaning\"\u003eCurrent breadth of participation\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Metric\"\u003eReparticipation rate\u003c/td\u003e\n\u003ctd data-label=\"Calculation or observation method\"\u003eMembers who participated during at least two periods ÷ participants in the previous period\u003c/td\u003e\n\u003ctd data-label=\"Meaning\"\u003eLevel of repeated participation\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Metric\"\u003eMember-to-member response rate\u003c/td\u003e\n\u003ctd data-label=\"Calculation or observation method\"\u003eQuestions answered by members ÷ all answered questions\u003c/td\u003e\n\u003ctd data-label=\"Meaning\"\u003eIndependence from the operator\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Metric\"\u003eResponse time\u003c/td\u003e\n\u003ctd data-label=\"Calculation or observation method\"\u003eMedian time from a question to the first valid answer\u003c/td\u003e\n\u003ctd data-label=\"Meaning\"\u003eSpeed of receiving help\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Metric\"\u003eParticipation concentration\u003c/td\u003e\n\u003ctd data-label=\"Calculation or observation method\"\u003eShare of messages from the most active participants\u003c/td\u003e\n\u003ctd data-label=\"Meaning\"\u003eRisk of dependence on a small group\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Metric\"\u003eNumber of member-led activities\u003c/td\u003e\n\u003ctd data-label=\"Calculation or observation method\"\u003eNumber of activities proposed and operated by members\u003c/td\u003e\n\u003ctd data-label=\"Meaning\"\u003eSelf-sustainability\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Metric\"\u003eReasons for leaving\u003c/td\u003e\n\u003ctd data-label=\"Calculation or observation method\"\u003eCategories identified through departure surveys or interviews\u003c/td\u003e\n\u003ctd data-label=\"Meaning\"\u003eClues for structural improvement\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\u003c/div\u003e\n\u003cp\u003eEven when message volume is high, satisfaction may be low if there is extensive small talk, duplicate notifications, or domination by the operator. Quantitative metrics should be supplemented with short interviews, anonymous surveys, and reviews of reasons for leaving.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#privacy-principles-for-analyzing-message-data\" class=\"anchor\" id=\"privacy-principles-for-analyzing-message-data\"\u003e\u003c/a\u003ePrivacy Principles for Analyzing Message Data\u003c/h2\u003e\n\u003cp\u003eCommunity conversations may contain information about participants’ relationships, interests, schedules, or personal concerns. The following principles are necessary when using them for operational analysis.\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eInform members of the purpose of collection and scope of analysis.\u003c/li\u003e\n\u003cli\u003eRemove names, contact details, and account identifiers that are unnecessary for the analysis.\u003c/li\u003e\n\u003cli\u003eDo not use private messages as public examples without consent.\u003c/li\u003e\n\u003cli\u003eLimit access to original messages and set a retention period.\u003c/li\u003e\n\u003cli\u003eBefore entering conversations into external AI tools, review the relevant service’s data-processing terms.\u003c/li\u003e\n\u003cli\u003ePresent public findings in aggregate units that do not allow individuals to be identified.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eThe platform’s terms of service and applicable privacy regulations must also be reviewed separately. The convenience of community analysis cannot take precedence over members’ privacy and trust.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#operating-principles-derived-from-the-nine-month-case\" class=\"anchor\" id=\"operating-principles-derived-from-the-nine-month-case\"\u003e\u003c/a\u003eOperating Principles Derived from the Nine-Month Case\u003c/h2\u003e\n\u003ch3\u003e\n\u003ca href=\"#start-small\" class=\"anchor\" id=\"start-small\"\u003e\u003c/a\u003eStart Small\u003c/h3\u003e\n\u003cp\u003eA community can be tested without a large follower base. Rather than gathering hundreds of people from the outset, securing 10 people who repeatedly converse and help one another makes it easier to learn how the operating structure works.\u003c/p\u003e\n\u003ch3\u003e\n\u003ca href=\"#assume-iterative-improvement-rather-than-a-finished-design\" class=\"anchor\" id=\"assume-iterative-improvement-rather-than-a-finished-design\"\u003e\u003c/a\u003eAssume Iterative Improvement Rather Than a Finished Design\u003c/h3\u003e\n\u003cp\u003eSign-up conditions, channel structure, roles, and inactivity criteria should be adjusted based on actual behavioral data. Explaining the reason for a change and its implementation date, then reviewing the results after a set period, can reduce arbitrary management.\u003c/p\u003e\n\u003ch3\u003e\n\u003ca href=\"#the-operator-is-a-designer-of-participation-structures-not-a-producer-of-conversations\" class=\"anchor\" id=\"the-operator-is-a-designer-of-participation-structures-not-a-producer-of-conversations\"\u003e\u003c/a\u003eThe Operator Is a Designer of Participation Structures, Not a Producer of Conversations\u003c/h3\u003e\n\u003cp\u003eInstead of answering every question first, the operator should give members time to respond. It is necessary to connect people with similar interests and give members the authority to execute activities they propose.\u003c/p\u003e\n\u003ch3\u003e\n\u003ca href=\"#examine-the-distribution-of-relationships-rather-than-the-total-number-of-people\" class=\"anchor\" id=\"examine-the-distribution-of-relationships-rather-than-the-total-number-of-people\"\u003e\u003c/a\u003eExamine the Distribution of Relationships Rather Than the Total Number of People\u003c/h3\u003e\n\u003cp\u003eIn addition to member count, assess how many people initiate conversations, whether different people provide answers, and whether core members are becoming exhausted. A healthy community has multiple connections rather than a single star operator.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#a-30-day-review-plan-for-implementation\" class=\"anchor\" id=\"a-30-day-review-plan-for-implementation\"\u003e\u003c/a\u003eA 30-Day Review Plan for Implementation\u003c/h2\u003e\n\u003cul\u003e\n\u003cli\u003e\n\u003cstrong\u003eWeek 1:\u003c/strong\u003e Define the community’s target audience, value provided, and permitted and prohibited behaviors in one sentence each.\u003c/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eWeek 2:\u003c/strong\u003e Document the journey from joining to the first introduction, first answer, and return visit, then identify drop-off points.\u003c/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eWeek 3:\u003c/strong\u003e Identify core members who help others and discuss their roles and burdens.\u003c/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eWeek 4:\u003c/strong\u003e Compare the operator’s message share, member-to-member responses, and member-led activities.\u003c/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eEnd of the month:\u003c/strong\u003e Decide which experiments to continue and which to stop, then explain the changes to members.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eThe key lesson from this case is that “gathering many people” and “helping people build relationships with one another” are different tasks. Community growth should be judged not only by a graph of member numbers, but also by whether questions, answers, and collaboration continue even when the operator temporarily steps back.\u003c/p\u003e\n","tags":["Creator","Community management","User engagement","Retention","MVP validation"],"faqs":[{"question":"Can I start a community even if I have few followers?","answer":"Yes. The initial goal suggested by this case is not to attract a large number of members, but to find a small group of members who participate repeatedly and help one another. However, regardless of follower count, the target audience, purpose, and basic rules must be clear."},{"question":"Can community membership and engagement be treated as the same metric?","answer":"No. Membership is the cumulative number of people who have joined the space, while engagement is the percentage of people who actually took action during a given period. It is better to look at the first-action rate, repeat participation rate, response rate among members, and number of member-led activities together."},{"question":"Does having many messages mean a community is healthy?","answer":"The number of messages alone is not enough to determine that. The operator or a small number of members may have written most of them, or there may be many duplicate notifications and casual conversations. The distribution of contributors, response quality, repeat participation, response time, and member satisfaction should be reviewed together."},{"question":"Why is it a problem if the operator starts many conversations?","answer":"It helps with initial engagement, but if every conversation depends on the operator, interactions may stop as soon as the operator takes a break. The operator should focus more on designing connections and rules that enable members to ask and answer one another's questions."},{"question":"Are core members the people who post the most messages?","answer":"Activity level is only one indicator. Healthy core members help new members, encourage others to speak, and respect the rules. High-frequency users who monopolize conversations or create conflict should be managed separately."},{"question":"Does requiring an application increase the participation rate?","answer":"It can help identify people with a clear purpose for participating, but it does not always increase the participation rate. If the questions are lengthy or request excessive personal information, even suitable people may drop out, so the application completion rate and post-joining activity rate should be compared together."},{"question":"Should all inactive members be removed?","answer":"Not necessarily. Members who only read can still gain information or participate later. If removal is necessary, the activity criteria and time period should be announced in advance, and members should be asked whether they wish to remain, with exceptions and a rejoining process provided."},{"question":"What does it mean for a community to operate organically?","answer":"It means that members answer questions, propose gatherings, and create value for one another without the operator having to provide a topic every time. However, organic operation does not mean a lack of management, and safety rules and a conflict resolution system remain necessary."},{"question":"Can MVP demand be validated solely through community feedback?","answer":"It is useful for qualitative feedback on the initial problem and proposed solution, but it is not sufficient. Active members may not represent the entire market, so external prospective customer research, actual sign-ups, and purchasing behavior after pricing is disclosed must also be assessed."},{"question":"Can community messages be analyzed with AI?","answer":"Participants should be informed of the purpose and scope, and identifying information must be removed. When entering original messages into an external AI service, the service's policies on storage, use for training, and deletion, as well as the platform's terms, should be reviewed, and sensitive conversations should not be transmitted without consent."}],"sources":[{"url":"https://www.nngroup.com/articles/participation-inequality/","title":"Participation Inequality: The 90-9-1 Rule for Social Features","type":"source"}],"images":[{"id":530,"url":"https://injoys.com/rails/active_storage/blobs/proxy/eyJfcmFpbHMiOnsiZGF0YSI6NjMyMSwicHVyIjoiYmxvYl9pZCJ9fQ==--6ad59d86034f3952759317ad7b7bc15095cbe16b/ai-213e9cfb.webp","is_representative":true,"generation_method":"ai_image","license":"ai_generated","mime_type":"image/webp","translations":{"ko":{"alt":"한 사람의 빈 네트워크가 여러 대화 그룹이 연결된 커뮤니티로 확장되는 일러스트","caption":"개별 크리에이터 네트워크가 활발한 커뮤니티로 성장하는 과정을 보여준다.","description":null},"en":{"alt":"A lone creator’s empty network expanding into connected groups in conversation","caption":"The illustration shows an individual creator network growing into an active community.","description":null},"ja":{"alt":"一人の空のネットワークが会話する複数のグループへ広がるイラスト","caption":"個人のクリエイターネットワークが活発なコミュニティへ成長する様子を表している。","description":null},"es":{"alt":"La red vacía de un creador se expande en varios grupos conectados que conversan","caption":"La ilustración muestra cómo una red individual crece hasta convertirse en una comunidad activa.","description":null},"id":{"alt":"Jaringan kosong seorang kreator berkembang menjadi kelompok percakapan yang saling terhubung","caption":"Ilustrasi ini menunjukkan jaringan kreator individu tumbuh menjadi komunitas yang aktif.","description":null},"pt":{"alt":"A rede vazia de um criador se expande em vários grupos conectados em conversa","caption":"A ilustração mostra uma rede individual crescendo e se tornando uma comunidade ativa.","description":null},"zh-hant":{"alt":"一名創作者的空白網絡擴展成多個彼此連結、互相交流的群組","caption":"插圖呈現個人創作者網絡逐步成長為活躍社群的過程。","description":null},"de":{"alt":"Das leere Netzwerk eines Kreativen wächst zu verbundenen Gesprächsgruppen","caption":"Die Illustration zeigt, wie aus einem individuellen Netzwerk eine aktive Community entsteht.","description":null}}},{"id":531,"url":"https://injoys.com/rails/active_storage/blobs/proxy/eyJfcmFpbHMiOnsiZGF0YSI6NjMyNywicHVyIjoiYmxvYl9pZCJ9fQ==--c41bce6ceb81e9b86d5c2fb4329839449dfaf08e/ai-07e7c37e.webp","is_representative":false,"generation_method":"ai_image","license":"ai_generated","mime_type":"image/webp","translations":{"ko":{"alt":"큰 아치를 지나 소규모 대화 모임과 커뮤니티로 합류하는 사람들","caption":"크리에이터들이 대화와 교류를 통해 연결되는 커뮤니티의 성장 과정을 보여준다.","description":null},"en":{"alt":"People passing through an arch to join connected community discussion groups","caption":"Creators build connections through conversations and small-group activities.","description":null},"ja":{"alt":"大きなアーチを通り、つながり合う少人数の交流グループに加わる人々","caption":"クリエイターが対話や少人数の活動を通じてつながる様子を表している。","description":null},"es":{"alt":"Personas cruzando un arco para unirse a grupos comunitarios de conversación","caption":"Los creadores establecen vínculos mediante conversaciones y actividades en grupos pequeños.","description":null},"id":{"alt":"Orang-orang melewati gerbang untuk bergabung dengan kelompok diskusi komunitas","caption":"Para kreator membangun koneksi melalui percakapan dan kegiatan kelompok kecil.","description":null},"pt":{"alt":"Pessoas atravessando um arco para participar de grupos comunitários conectados","caption":"Criadores formam conexões por meio de conversas e atividades em pequenos grupos.","description":null},"zh-hant":{"alt":"人們穿過拱門，加入彼此連結的社群討論小組","caption":"創作者透過對話與小組活動建立連結。","description":null},"de":{"alt":"Menschen gehen durch einen Torbogen zu vernetzten Gesprächsgruppen einer Community","caption":"Kreative knüpfen durch Gespräche und Aktivitäten in kleinen Gruppen neue Kontakte.","description":null}}}],"published_at":"2026-08-08T04:03:26+09:00","updated_at":"2026-08-08T04:03:26+09:00","license":"cc_by","translation_status":"reviewed","available_locales":["ko","en","ja","es"],"data_locales":["ko","en","ja","es","id","pt","zh-hant","de"],"url":"https://injoys.com/en/articles/creator-community-nine-month-operation-case-study"}