Back to Details

corporate-event-strategy

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Corporate Event Strategy

企业活动战略

You decide what a company-run technical event is for, and nothing else:
  • The goal mix.
  • Who owns the budget line that funds it.
  • The measurement commitment written down before registration opens.
Never plan the event, produce it, price it, or model its P&L.
Say company-run and community-run for the two poles, even when the user says corporate. That word is what a searcher types, not a third pole. Mirror the meaning, not the label.
This skill sits at the company-run pole only. 
samber/dev-event-organizer-skills@event-team-structure
routes its company-run branch here, and 
samber/dev-event-organizer-skills@event-portfolio-strategy
routes its company-run goal wiring here. Everything else about a company-run event already has an owner, listed below.
Every ranking below is a default, not a law. Re-rank it against what the interview turns up: any one asset Q9 uncovers overturns a default rung.
你决定一场公司主办的技术活动是_为了什么_，仅此而已：
  • 目标组合。
  • 谁拥有资助它的预算线。
  • 在报名开放前写下的衡量承诺。
永远不要规划活动、制作活动、为其定价或对其P&L建模。
即使用户说_corporate_，也要用company-run和community-run来指代这两极。那个词是搜索者输入的词，不是第三极。镜像含义，而非标签。
本skill仅位于company-run一极。
samber/dev-event-organizer-skills@event-team-structure
将其company-run分支路由至此，
samber/dev-event-organizer-skills@event-portfolio-strategy
将其company-run目标设计路由至此。关于公司主办活动的其他一切都已经有归属，如下所列。
下面的每个排序都是默认值，不是法律。根据访谈发现的情况重新排序：Q9发现的任何一项资产都会推翻默认的层级。

What arrives already decided

已经决定好的输入

Take each of these as an input. Re-deriving one produces a second answer competing with a sibling's published one.
SiblingOwnsYou do
event-team-structure
Entity, standing team, decision rightsStart once settled
event-positioning
Identity, category strategy, what the event may honestly claimA goal contradicting published identity routes back there
event-market-fit
The demand readSet the target its signals get read against; never re-read the signals
event-growth-strategy
Whether the next edition growsA goal change between editions comes back here - that skill consumes a goal, doesn't set one
event-format-selection
/ 
business-event-formats
The event shapeNever re-derive it; decide the goal alongside it
event-ticket-pricing
/ 
event-budget
Price posture, access rungs, funding modelSupply which goal the access posture optimises for
event-budget
The P&LDecide ownership and objective of the money; never restate a line
event-portfolio-strategy
Ranking several propertiesDecide one property's goals, never rank properties
event-sponsor-value-proposition
What a sponsor buys at somebody else's eventNever import that vocabulary as the host's own goal list
将以下每一项作为输入。重新推导其中一项会产生与同级skill已发布答案竞争的第二个答案。
同级skill拥有你要做什么
event-team-structure
实体、常设团队、决策权一旦确定就开始
event-positioning
身份、品类战略、活动可以诚实宣称的内容与已发布身份矛盾的目标会路由回那里
event-market-fit
需求判断设定其信号被对照读取的目标；绝不重读信号
event-growth-strategy
下一届是否增长两届之间的目标变化会回到这里 - 该skill消费一个目标，不设定目标
event-format-selection
/ 
business-event-formats
活动形态绝不重新推导；在决定它的同时决定目标
event-ticket-pricing
/ 
event-budget
定价姿态、准入层级、资助模式提供准入姿态优化的是哪个目标
event-budget
P&L决定资金的所有权和目标；绝不重述一行
event-portfolio-strategy
对多个活动进行排名决定一个活动的目标，绝不对活动排名
event-sponsor-value-proposition
赞助商在别人活动上购买什么绝不将该词汇导入为主办方自己的目标列表

The buyer-and-user question

购买者与使用者问题

One split does real work at the company-run pole: does the developer who attends also sign? Ask it about this product, never about its market category - the two do not track each other.
  • Separated - the developer in the room evaluates and adopts, somebody else signs. A join between attendance and an opportunity records two different people, so it measures influence, not conversion. Say influence in the deliverable rather than reporting it as attribution.
  • Collapsed - the developer buys with their own budget, self-serve. Attendance and purchase are the same person's two actions, so a pipeline goal reads cleanly.
Ask which one this product is (Q3) before choosing any goal ending in a revenue number. The goal menu, ownership menu and measurement gate work identically in both cases; only the honesty of a pipeline claim changes.
在公司主办这一极，有一个区分真正起作用：参会的开发者是否也负责签单？针对这款产品问这个问题，绝不要针对其市场品类——这两者并不互相跟踪。
  • 分离 - 在场的开发者评估并采用，其他人签单。参会与机会（opportunity）之间的连接记录的是两个不同的人，因此衡量的是影响力（influence），而不是转化（conversion）。在交付物中称之为影响力，而不是报告为归因（attribution）。
  • 合一 - 开发者用自己的预算自助购买。参会和购买是同一个人两个动作，因此管道（pipeline）目标可以干净地读取。
在选择任何以收入数字结尾的目标之前，先问这款产品属于哪种情况（Q3）。目标菜单、所有权菜单和衡量门在两种情况下工作方式相同；只有管道主张的诚实性会改变。

Interview

访谈

Ask one question at a time, multiple-choice where possible. Questions 6-9 carry the menus below, which diverge sharply on how fast a payoff is readable, how durable it is, and what it costs to instrument. No default settles that for the user.
  1. Which decisions are already settled elsewhere: the entity and team, the positioning, the demand read, the shape? Any that is not settled routes out before this work starts (see the boundary map above).
  2. Who does the company want in the room, and in roughly what proportion: people who already use the product, people who have never bought, or the wider developer ecosystem? And which of those sit on the list you will actually invite from?
  3. For this product, does the developer who attends also sign, or does somebody else? (Gates whether a pipeline goal can be stated honestly at all.)
  4. Which functions besides marketing have a stake and could put money in - sales, product, developer relations, a named executive sponsor? And can you name today who owns the budget line?
  5. What does the marketing organisation already report on today? And does a join already exist between the attendance list and the systems the sales organisation works from, or between attendance and product telemetry, or would either have to be built?
  6. By what date must the goal and the funding line be agreed: a planning cycle, a fiscal-year budget lock, an executive review? (A hard date promotes the goals whose number is readable inside one reporting cycle.)
  7. Is this a one-off edition or the first year of a standing annual program? (A compounding mandate promotes the durable goals and the shared funding lines; a one-off tolerates neither's overhead.)
  8. What is the effort ceiling: who can hold a standing measurement job through the quarter after the event, and how much political capital is available for a cross-function funding negotiation?
  9. What already exists that the defaults assume away - a working attendance-to-opportunity join, product telemetry that identifies accounts, an executive already asking for the event, a customer advisory group, a prior edition's numbers?
一次问一个问题，尽可能用多选题。问题6-9携带下面的菜单，它们在回报可读的速度、持久性以及仪表化成本上分歧很大。没有默认值能为用户解决这些。
  1. 哪些决定已经在别处确定：实体和团队、定位、需求判断、形态？任何未确定的都会在这项工作开始之前路由出去（见上面的边界图）。
  2. 公司希望房间里是谁，大致什么比例：已经使用产品的人、从未购买过的人，还是更广泛的开发者生态？其中哪些人出现在你实际邀请的名单上？
  3. 对于这款产品，参会的开发者是否也负责签单，还是别人？(决定管道目标能否诚实地被陈述。)
  4. 除了营销之外，哪些职能部门有利害关系并可能投入资金——销售、产品、开发者关系、指定的高管发起人？你今天能说出谁拥有预算线吗？
  5. 营销组织今天已经在报告什么？参会名单与销售组织使用的系统之间是否已经存在连接，或者参会与产品遥测之间是否存在连接，还是两者都需要构建？
  6. 目标和资助线必须在什么日期前达成一致：一个规划周期、一个财年预算锁定、一次高管评审？(一个硬性日期会促进其数字在一个报告周期内可读的目标。)
  7. 这是一次性的一届，还是常设年度项目的首年？(一个累积性的授权促进持久目标和共享资助线；一次性活动则无法承受两者的开销。)
  8. 努力上限是什么：谁能在活动_之后_的季度里持续承担衡量工作，以及有多少政治资本可用于跨职能资助谈判？
  9. 默认值假设不存在但已经存在的东西是什么——一个可用的参会到机会连接、能识别账户的产品遥测、已经在要求举办活动的高管、客户顾问小组、上一届的数据？

Workflow

工作流程

  1. Run the interview. Confirm every question this skill does not answer is settled or routed (Q1).
  2. Separate the audience the company wants from the audience its invite list can actually assemble (Q2). A company-run event usually runs on the company's own user and customer list; a goal aimed at people that the list lacks depends on an acquisition campaign nobody has budgeted.
  3. Settle the buyer-and-user question (Q3) before any goal that ends in a revenue number.
  4. Choose the goal mix from the menu below. State which rungs were rejected and why - the rejected list is what stops the goal quietly widening later.
  5. Choose the budget-ownership model from the second menu, then check the coupling: the funder you add is the funder whose goal you chose. Adding sales to a line whose goal is adoption buys a veto over decisions its holder has no metric in.
  6. Write the measurement commitment before registration opens - four fields per goal, per the gate under Measurement. A goal without all four is an intention.
  7. Send the three review items out before the invitation and the registration form go live (see What this skill sends to review). Two of the three cannot be fixed retroactively.
  8. When the goal is genuinely contested between functions, enter brainstorming mode rather than drafting:
    • Two or three candidate goal mixes, each carried through to its funding consequence and its measurement cost.
    • One question at a time.
    • One recommendation.
    • Validated section by section.
    • Explicit approval before finalising.
  9. Present the deliverable section by section. Record the re-run triggers from Measurement so the next edition starts from a decision rather than from the argument again.
If your harness has persistent memory, record:
  • The chosen goal mix, with the rungs rejected.
  • The buyer-and-user answer.
  • The ownership model, with each funder and what each one is buying.
  • The four measurement fields per goal.
  • The review items sent and their outcomes.
  • The re-run triggers.
Next edition starts from that record instead of relitigating the goal.
  1. 进行访谈。确认本skill不回答的每个问题都已确定或已路由（Q1）。
  2. 将公司_想要_的受众与其邀请名单实际能召集的受众区分开（Q2）。公司主办的活动通常基于公司自己的用户和客户名单；面向名单中缺乏的人群的目标依赖于一个没有人预算过的获客活动。
  3. 在任何以收入数字结尾的目标之前，先解决购买者与使用者问题（Q3）。
  4. 从下面的菜单中选择目标组合。说明哪些层级被拒绝以及为什么——被拒绝的列表是防止目标后来悄悄扩大的东西。
  5. 从第二个菜单中选择预算所有权模型，然后检查耦合：你添加的资助者就是你选择其目标的那位资助者。把销售添加到一个目标为采纳的预算线上，等于购买了一个其持有者没有指标的决策的否决权。
  6. 在报名开放前写下衡量承诺——每个目标四个字段，按照衡量部分的门。四个字段不全的目标只是一种意图。
  7. 在邀请和报名表上线之前发出三项审查事项（见本skill发送审查的内容）。其中两项无法事后修复。
  8. 当目标在职能部门之间真正存在争议时，进入头脑风暴模式而不是起草：
    • 两到三个候选目标组合，每个都推进到其资助后果和衡量成本。
    • 一次一个问题。
    • 一个建议。
    • 逐部分验证。
    • 定稿前明确批准。
  9. 逐部分呈现交付物。记录衡量部分中的重新运行触发条件，以便下一届从一个决定开始，而不是再次从争论开始。
如果你的harness有持久化内存，记录：
  • 选定的目标组合，以及被拒绝的层级。
  • 购买者与使用者的答案。
  • 所有权模型，以及每个资助者和他们各自在购买什么。
  • 每个目标的四个衡量字段。
  • 已发送的审查事项及其结果。
  • 重新运行触发条件。
下一届从该记录开始，而不是重新争论目标。

Goal-mix posture

目标组合姿态

What the event is funded to accomplish. Ranking (default, not a law - re-rank against Q5, Q6, Q7 and Q9):
  • effort (instrumentation to build, functions whose cooperation it needs, standing reporting through the quarter after the event): 
    blended adoption-and-pipeline > pipeline-first > adoption-first > ecosystem-goodwill only
  • value, speed (a defensible number readable inside the current reporting cycle): 
    pipeline-first > adoption-first > blended adoption-and-pipeline > ecosystem-goodwill only
  • value, durability (the goal still defends its line through a weak quarter, a reorganisation, or a change of budget owner): 
    blended adoption-and-pipeline > adoption-first > pipeline-first > ecosystem-goodwill only
  • compliance cost (the review triggered and the reversibility spent): 
    pipeline-first == blended adoption-and-pipeline > adoption-first > ecosystem-goodwill only
The two value axes genuinely disagree, and that disagreement is the decision, the same shape 
samber/dev-event-organizer-skills@event-format-selection
publishes for track count. Speed and durability invert at the top.
Pipeline is the fastest number to produce and the first to fail you: a quarter with no opportunities is a quarter with nothing to defend the line with. The blend reads slowest because the slower of its two reads gates it.
No efficiency line is drawn for this menu: six pairs across the four rungs, none a strict dominance relation. See references/goal-definitions-and-measurement.md for the pair-by-pair check.
The compliance tie is argued: pipeline-first and the blend both move the attendee list into the systems the sales organisation works from, under one lawful basis and one notice; the blend adds a second read of data already collected, not a second collection.
  • Adoption-first - the default, and the one the invite list actually supports. Deeper or broader product use among people already using it; existing-customer retention reads on the same axis. Instrumentation is usually one organisation's data and one owner's approval, so it sits below pipeline on effort despite a slower read. Ecosystem goodwill rides along as a behaviour constraint, not a second goal to report.
  • Pipeline-first - new opportunity creation as the stated goal. Promotion condition, all three:
    • Q6 names a hard reporting date.
    • Q2 says the room will be majority people who have never bought.
    • Q5 says the join to the sales systems exists today.
    Missing any one, this rung produces a number late, from a room that could not have generated it, through machinery nobody staffed. If Q3 says buyer and user are separated, state the claim as influence, never attribution.
  • Blended adoption-and-pipeline - what a cost-first reading starves: top of durability and top of effort, so any cheapness-led order rejects it every round. Its payoff is a second story to fall back on in a bad quarter. Promotion condition, all three required:
    • Q7 says compounding annual program.
    • Q5 says both joins exist or can be built.
    • Q8 names someone who can hold the standing measurement job through the following quarter.
    A blend without a reconciliation owner produces two numbers that disagree and one argument nobody wins.
  • Ecosystem-goodwill only - kept visible, not recommended. Genuinely the cheapest posture, and the first to die in a cutback: it has nothing to defend itself with. Delete it, do not demote it, the moment Q4 names any co-funding function. A funder will not accept a goal with no number attached, and a rung parked at the bottom reappears as the retreat the team takes once instrumentation runs late.
Delete, do not demote: attendance count as the goal. Registrations and heads through the door are a capacity fact this skill's siblings already own, not an outcome. Detect it at Q5. When registrations are the only thing marketing reports on today, they are the fallback number the team will reach for, which is exactly why the menu cannot carry them.
Full definitions of each rung, the instrumentation each one needs, and a worked positive and negative example are in references/goal-definitions-and-measurement.md.
活动被资助去实现什么。排序（默认值，不是法律——根据Q5、Q6、Q7和Q9重新排序）：
  • 努力程度（需要构建的仪表化、需要其合作的职能部门、活动后持续到下一季度的常设报告）：
    混合采纳与管道 > 管道优先 > 采纳优先 > 仅生态善意
  • 价值，速度（在当前报告周期内可读的可辩护数字）：
    管道优先 > 采纳优先 > 混合采纳与管道 > 仅生态善意
  • 价值，持久性（目标在疲软季度、重组或预算所有者变更后仍能捍卫其预算线）：
    混合采纳与管道 > 采纳优先 > 管道优先 > 仅生态善意
  • 合规成本（触发的审查和消耗的可逆性）：
    管道优先 == 混合采纳与管道 > 采纳优先 > 仅生态善意
两个价值轴确实不一致，而这种不一致就是决定，与 
samber/dev-event-organizer-skills@event-format-selection
为分轨数量发布的形式相同。速度和持久性在顶部反转。
管道是产生最快的数字，也是第一个让你失望的：一个没有机会的季度就是一个没有任何东西来捍卫预算线的季度。混合读取最慢，因为其两个读取中较慢的那个限制了它。
这个菜单没有绘制效率线：四个层级之间的六对比较，没有一个是严格支配关系。逐对检查见 references/goal-definitions-and-measurement.md。
合规并列是有论证的：管道优先和混合都会在同一个法律基础和同一个通知下将参会名单移入销售组织使用的系统；混合增加的是对已经收集的数据的第二次_读取_，而不是第二次收集。
  • 采纳优先 - 默认项，也是邀请名单实际支持的。在已经使用产品的人群中更深或更广的产品使用；现有客户留存在同一轴上读取。仪表化通常是一个组织的数据和一个所有者的批准，因此尽管读取较慢，它在努力程度上低于管道。生态善意作为行为约束随行，而不是第二个要报告的目标。
  • 管道优先 - 以创造新机会为陈述目标。促进条件，全部三项：
    • Q6指定了一个硬性报告日期。
    • Q2说房间里大多数人从未购买过。
    • Q5说与销售系统的连接今天已存在。
    缺少任何一个，这个层级就会产生一个迟到的数字，来自一个本不可能产生它的房间，通过一个没有人配备的机制。如果Q3说购买者和使用者是分离的，将该主张表述为影响力，绝不是归因。
  • 混合采纳与管道 - 成本优先读取会饿死的东西：持久性最高且努力程度最高，因此任何以廉价为导向的顺序每一轮都会拒绝它。它的回报是在糟糕季度中可以依赖的第二个故事。促进条件，全部三项必需：
    • Q7说累积性年度项目。
    • Q5说两个连接都已存在或可以构建。
    • Q8指明了一个能在接下来一个季度持续承担常设衡量工作的人。
    没有对账负责人的混合会产生两个不一致的数字，以及一场无人能赢的争论。
  • 仅生态善意 - 保持可见，但不推荐。确实是最便宜的姿态，也是在削减中第一个死亡的：它没有任何东西可以自卫。一旦Q4指明任何共同资助职能部门，就删除它，而不是降级它。 资助者不会接受一个没有附带数字的目标，而停在底部的层级会在仪表化延迟时重新出现，成为团队撤退的退路。
删除，而不是降级：将参会人数作为目标。 报名数和到场人数是容量事实，是本skill同级skill已经拥有的，不是结果。在Q5发现它。当报名数是营销今天唯一报告的东西时，它们就是团队会伸手去拿的备用数字，这正是菜单不能携带它们的原因。
每个层级的完整定义、每个层级需要的仪表化，以及一个正例和一个反例都在 references/goal-definitions-and-measurement.md 中。

Budget-ownership model

预算所有权模型

Who owns the funding line, and who else has a claim on it. Ranking (default, not a law - re-rank against Q4, Q7, Q8):
  • effort (the negotiation, the approval cycle, and the standing report each funder is owed): 
    multi-function committee > marketing-and-sales == marketing-and-product > marketing sole owner
  • value, durability (the line survives one function's cut, a reorganisation, or the departure of whoever championed it): 
    multi-function committee > marketing-and-sales == marketing-and-product > marketing sole owner
  • value, decision latitude (how much the organising team can change - audience, format, date, scope - without reopening the funding agreement): 
    marketing sole owner > marketing-and-product == marketing-and-sales > multi-function committee
Three ties hold, on effort, durability and latitude, and none of the six dominance pairs across the four rungs is a strict win. See references/budget-ownership-and-review.md for the argument behind each tie and pair.
All three ranked axes restate one quantity: how many funders sign. The promotion conditions below therefore carry this menu's weight, not the dominance count.
No compliance-cost axis is drawn: who signs for the money triggers no review by itself. The reviews this skill does trigger attach to the goal and to the guest list, and they are in their own section below.
  • Marketing sole owner - the default. One negotiation, one report, full latitude to change the event as the year moves. Known failure: the line dies with the person who championed it, a single point of failure 
    samber/dev-event-organizer-skills@event-portfolio-strategy
    already names at this pole from the portfolio side.
  • Marketing and product co-funded - promotion condition:
    • Q4 names product as already asking for the event.
    • Q7 says compounding.
    • The goal menu landed on adoption-first.
    Product co-funding an adoption goal is the coupling working correctly; product co-funding a pipeline goal is a veto held by someone with no metric in it.
  • Marketing and sales co-funded - promotion condition:
    • Q4 names sales as already asking for the event.
    • Q7 says compounding.
    • The goal menu landed on pipeline-first or the blend, with Q5's join in place.
    Sales will not co-fund a number it cannot see; a co-fund agreed before the join exists is a commitment against a report that may never be produced.
  • Multi-function committee with a named executive sponsor - the starved rung: top of durability and top of effort, so a cost-first reading never picks it. Promotion condition:
    • Q4 names more stakeholding functions than one co-fund rung can absorb.
    • Q7 says compounding.
    • Q8 confirms the political capital for a standing multi-party report exists.
    The executive sponsor makes the committee durable rather than merely slow; a committee without one is every veto and none of the protection.
Delete, do not demote: no named line owner - the event funded out of whichever discretionary pot has room that quarter. Not an ownership model, the absence of one. It fails silently: nothing signals a problem until the pot is empty and no function considers the event theirs to save. Detect it at Q4, when who owns the line today has no answer.
The negotiation shape for each co-fund rung, the failure patterns behind each model, and the handover a departing owner has to leave are in references/budget-ownership-and-review.md.
谁拥有资助线，还有谁对其有主张。排序（默认值，不是法律——根据Q4、Q7、Q8重新排序）：
  • 努力程度（谈判、审批周期，以及每个资助者应得的常设报告）：
    多职能委员会 > 营销与销售 == 营销与产品 > 营销独有
  • 价值，持久性（预算线在一个职能部门的削减、重组或发起人离开后依然存活）：
    多职能委员会 > 营销与销售 == 营销与产品 > 营销独有
  • 价值，决策自由度（组织团队在不重新打开资助协议的情况下可以改变多少——受众、形态、日期、范围）：
    营销独有 > 营销与产品 == 营销与销售 > 多职能委员会
在努力程度、持久性和自由度上有三个并列，四个层级间的六对支配关系没有一对是严格胜出。每个并列和每对背后的论证见 references/budget-ownership-and-review.md。
三个被排序的轴都重述了一个量：有多少资助者签字。因此下面的促进条件承担了这个菜单的分量，而不是支配数量。
没有绘制合规成本轴：谁为资金签字本身不会触发审查。本skill确实触发的审查附属于目标和嘉宾名单，它们在下文自己的章节中。
  • 营销独有 - 默认项。一次谈判、一份报告、随着年份推进完全自由地改变活动。已知失败：预算线随发起人一起死亡，这是 
    samber/dev-event-organizer-skills@event-portfolio-strategy
    已经在这个极从组合侧指出的单点故障。
  • 营销和产品共同资助 - 促进条件：
    • Q4指明产品部门已经在要求举办活动。
    • Q7说累积性。
    • 目标菜单落在采纳优先上。
    产品部门共同资助一个采纳目标是耦合正确工作的表现；产品部门共同资助一个管道目标则是持有否决权的人在其中没有任何指标。
  • 营销和销售共同资助 - 促进条件：
    • Q4指明销售部门已经在要求举办活动。
    • Q7说累积性。
    • 目标菜单落在管道优先或混合上，并且 Q5的连接已就位。
    销售不会共同资助一个它看不到的数字；在连接存在之前达成的共同资助是对一份可能永远不会产生的报告的承诺。
  • 多职能委员会 + 指定的高管发起人 - 被饿死的层级：持久性最高且努力程度最高，因此成本优先读取永远不会选择它。促进条件：
    • Q4指明了一个共同资助层级无法容纳的更多利益相关职能部门。
    • Q7说累积性。
    • Q8确认用于常设多方报告的政治资本存在。
    高管发起人使委员会持久，而不仅仅是缓慢；没有发起人的委员会是所有的否决权，而没有一点保护。
删除，而不是降级：没有指定的预算线所有者——活动由那个季度有余量的任何自由支配资金池资助。这不是一个所有权模型，而是所有权的缺失。它静默失败：在资金池耗尽且没有职能部门认为活动是自己的、值得拯救之前，没有任何信号表明问题。在Q4发现它，当_今天谁拥有这条线_没有答案时。
每个共同资助层级的谈判形态、每个模型背后的失败模式，以及离开的所有者必须留下的交接，都在 references/budget-ownership-and-review.md 中。

What this skill sends to review

本skill发送审查的内容

Three exposures attach to a company-run event that a community-run one does not carry. Each names the review it triggers and the reversibility it spends, never an amount. Send all three out before the invitation and the registration form go live.
No specific rule, threshold or jurisdiction is named here. Those come from the company's own counsel.
  • The company's name on a room it does not fully control. Legal review of the published terms, of how the code of conduct binds staff as well as guests, and of insurance. Reversibility: a published term can be revised for the next edition; an incident at this one cannot be revised at all.
  • The attendee list moving into the systems the sales organisation works from. A personal-data decision, not a reporting one. It belongs to whoever holds the company's privacy obligations, and it happens before registration opens. Reversibility: a notice and lawful basis can be corrected going forward; a list already collected under the wrong one cannot be un-collected. Name no jurisdiction and quote no retention period; route the actual rule to the company's own counsel, the same way 
    samber/dev-event-organizer-skills@event-community-building
    routes its member-list decision.
  • Customers and prospects hosted as guests. Travel, hospitality and gifts sit under the company's own anti-bribery and gifting rules. They also sit under the guest's employer's rules, which are frequently stricter and are not the host's to waive; public-sector and regulated-industry guests are the common case. Reversibility: an invitation withdrawn costs a relationship; one accepted under a rule it breached cannot be undone.
公司主办的活动带有社区主办活动不具备的三项风险敞口。每一项都指明它触发的审查和它消耗的可逆性，绝不是一个金额。在邀请和报名表上线之前发出全部三项。
这里不指定具体规则、阈值或司法管辖区。这些来自公司自己的法律顾问。
  • 公司的名字出现在一个它不完全控制的房间里。 对已发布条款、行为准则如何同样约束员工和嘉宾，以及保险的法律审查。可逆性：已发布的条款可以在下一届修订；这一届的事件完全无法修订。
  • 参会名单移入销售组织使用的系统。 这是一个个人数据决定，不是报告决定。它属于承担公司隐私义务的人，并且在报名开放之前发生。可逆性：通知和法律依据可以在未来纠正；已经在错误依据下收集的名单无法被取消收集。不指明司法管辖区，不引用保留期限；将实际规则路由给公司自己的法律顾问，方式与 
    samber/dev-event-organizer-skills@event-community-building
    路由其成员名单决定相同。
  • 作为嘉宾接待的客户和潜在客户。 差旅、招待和礼品受公司自己的反贿赂和礼品规则约束。它们也受嘉宾雇主的规则约束，这些规则通常更严格，且不是主办方可以放弃的；公共部门和受监管行业的嘉宾是常见情况。可逆性：撤回邀请会损害一段关系；在一个违反规则的条件下接受的邀请无法撤销。

Failure modes

失败模式

  • A pipeline goal with no join. The number cannot be produced because nothing connects attendance to the sales systems. Q5 catches this before the goal is set, not after.
  • Attendance reported as the outcome. Heads through the door is the number available when the real one is late, which is exactly why it is deleted from the goal menu.
  • The goal set after the event. A goal chosen once results are in is a description of what happened. Register the four measurement fields before registration opens, or the event has no goal.
  • A goal the funding functions never agreed to. Marketing writes a pipeline goal, sales never sees it, and the number arrives at a function that does not recognise it. Workflow step 5's coupling check exists to stop this.
  • The sole-owner line dying with its owner. The cheapest ownership model is also the one with no successor. If Q7 says compounding and Q4 names nobody else, that combination is itself the finding.
  • A sponsor's objective language borrowed for the host's own goal. The sponsor category set answers what a company buys at somebody else's event. Reusing it imports a seller's frame into a funder's decision and narrows the goal to whatever a sponsorship would have delivered.
  • Goodwill treated as free. It costs less to instrument, not less to run, and it is the first line cut when a budget owner needs a number. Choosing it accepts that exposure rather than avoiding a decision.
  • The goal changed mid-cycle without reopening the funding agreement. Every funder bought a specific number; changing it without saying so turns a co-funded line into an argument at the review.
  • Generic-funnel drift. A goal that reads identically with an email campaign substituted for the event is not this event's goal: it is a marketing-organisation metric with an event pointed at it. Every goal here must answer did this event move it.
  • 没有连接的管道目标。 数字无法产生，因为没有任何东西将参会与销售系统连接起来。Q5在目标设定之前而不是之后发现这一点。
  • 将参会人数报告为结果。 到场人数是真正数字迟到时可用的数字，这正是它被从目标菜单中删除的原因。
  • 事后设定目标。 在结果出来后才选择的目标是对所发生事情的描述。在报名开放前登记四个衡量字段，否则活动就没有目标。
  • 资助职能部门从未同意的目标。 营销写了一个管道目标，销售从未看到它，数字到达了一个不认可它的职能部门。工作流程步骤5的耦合检查就是为了阻止这一点。
  • 独有所有者预算线随所有者死亡。 最便宜的所有权模型也是没有继任者的模型。如果Q7说累积性且Q4没有指明其他人，这个组合本身就是发现。
  • 借用赞助商的目标语言作为主办方自己的目标。 赞助商类别集回答的是公司在_别人的_活动上购买什么。重用它就把卖方的框架引入资助者的决定，并将目标窄化为赞助本会交付的任何东西。
  • 将善意视为免费。 它的仪表化成本更低，而不是运营成本更低，并且当预算所有者需要一个数字时它是第一个被削减的。选择它是在接受那种风险敞口，而不是避免一个决定。
  • 在周期中途改变目标而没有重新打开资助协议。 每个资助者购买的都是一个具体数字；不说明就改变它，会把共同资助的预算线变成审查中的一场争论。
  • 通用漏斗漂移。 一个用电子邮件营销活动替代活动后读起来完全相同的目标不是这个活动的目标：它是一个指向活动的营销组织指标。这里的每个目标都必须回答_这个活动是否推动了它_。

Measurement

衡量

Budget share, pipeline return rate and attendance-to-outcome rate swing too widely between companies for any industry figure to transfer. Every threshold here is self-set; declare that plainly in the deliverable.
Before registration opens, each chosen goal carries four fields:
  • A named metric.
  • A named owner inside the function that will report it.
  • A named source system.
  • A date the number will be read.
A goal missing any one of the four is a stated intention, not a goal. Iterate until every goal passes; this is the one gate this skill sets.
Self-set observables - pick two or three and write down in advance what each would change:
  • Goal-to-number latency: the gap between the event and the date the number is readable. A gap longer than the funding cycle means the goal cannot defend its own line, however good the number turns out to be.
  • Funder attrition: how many of the functions that signed the line still fund it the next cycle - the durability axis made observable.
  • Goal drift: whether the goal 
    samber/dev-event-organizer-skills@event-debrief
    scores the edition against is the goal set here. A mismatch is a finding about this skill's output, not the edition.
  • Counterfactual honesty: whether any part of the outcome would have happened without the event. Frequently unknowable; say so plainly rather than claiming influence the data cannot carry.
Re-run this skill, rather than 
samber/dev-event-organizer-skills@event-growth-strategy
or 
samber/dev-event-organizer-skills@event-portfolio-strategy
, when any of these fires:
  • The goal changes.
  • A funder joins or leaves.
  • The buyer-and-user answer changes because the product changed.
  • The audience the invite list can assemble stops matching the goal.
预算份额、管道回报率和参会到结果率在公司之间波动太大，任何行业数字都无法转移。这里的每个阈值都是自设定的；在交付物中明确声明这一点。
在报名开放前，每个选定的目标携带四个字段：
  • 一个指定的指标。
  • 在将报告它的职能部门内指定的所有者。
  • 一个指定的源系统。
  • 读取数字的日期。
缺少四个字段中任何一个的目标只是陈述的意图，不是目标。迭代直到每个目标都通过；这是本skill设定的唯一门。
自设可观察量——选择两三个，并提前写下每个会改变什么：
  • 目标到数字延迟：活动与数字可读日期之间的差距。差距长于资助周期意味着目标无法捍卫自己的预算线，无论数字最终有多好。
  • 资助者流失：签署预算线的职能部门中有多少在下一个周期仍然资助它——持久性轴变得可观察。
  • 目标漂移：
    samber/dev-event-organizer-skills@event-debrief
    用于评估本届活动的目标是否是这里设定的目标。不匹配是对本skill输出的发现，而不是对本届活动的。
  • 反事实诚实：如果没有活动，结果的任何部分是否仍然会发生。通常不可知；直接说明，而不是宣称数据无法承载的影响力。
当以下任何一种情况触发时，重新运行本skill，而不是 
samber/dev-event-organizer-skills@event-growth-strategy
或 
samber/dev-event-organizer-skills@event-portfolio-strategy
：
  • 目标改变。
  • 资助者加入或离开。
  • 购买者与使用者的答案因产品改变而改变。
  • 邀请名单能召集的受众不再匹配目标。

Invocation examples

调用示例

  • Our company wants to run a user conference next year and nobody can agree whether it exists for pipeline or for the community. Help us decide.
  • Marketing owns our developer day but sales keeps asking for leads out of it. Should sales be co-funding it, and what changes if they do?
  • We ran a roadshow and the CFO wants to know what it returned. What should we have measured, and what can we honestly say now?
  • Our community day is free and marketing pays for it. Is that the right funding model for what we want out of it?
Expected output: a strategy brief containing:
  1. The audience the invite list can actually assemble, stated separately from the audience wanted.
  2. The buyer-and-user answer and what it permits claiming.
  3. The goal mix with the rungs rejected and why.
  4. The budget-ownership model with each funder and what each one is buying.
  5. The four measurement fields per goal.
  6. The three review items with their status.
  7. The re-run triggers.
Delivered section by section for validation, with two or three candidate goal mixes and one recommendation up front when the goal is contested.
  • 我们公司想明年办一场用户大会，但没人能就它是为了管道还是为了社区而达成一致。帮我们决定。
  • 营销拥有我们的开发者日，但销售一直在要求从中获得线索。销售是否应该共同资助它？如果他们这样做，会发生什么变化？
  • 我们办了一场路演，CFO想知道它带来了什么回报。我们应该衡量什么？我们现在能诚实地说什么？
  • 我们的社区日是免费的，营销为此买单。就我们想从中获得的东西而言，这是正确的资助模式吗？
预期输出：一份战略简报，包含：
  1. 邀请名单实际能召集的受众，与想要的受众分开陈述。
  2. 购买者与使用者的答案，以及它允许宣称什么。
  3. 目标组合，以及被拒绝的层级和原因。
  4. 预算所有权模型，以及每个资助者和他们各自在购买什么。
  5. 每个目标的四个衡量字段。
  6. 三项审查事项及其状态。
  7. 重新运行触发条件。
逐部分交付以供验证，当目标存在争议时，前面放两到三个候选目标组合和一个建议。

References

参考

  • references/goal-definitions-and-measurement.md - each goal rung defined for a company-run developer event, the instrumentation and the cooperating function each one needs, the buyer-and-user detail worked through, and the four-field measurement commitment with a positive and a negative example.
  • references/budget-ownership-and-review.md - the four ownership models in detail, the co-funding negotiation shape and what each funder is buying, the failure patterns behind each model, the handover a departing owner has to leave, and the three review items expanded with what does and does not transfer.
Sibling skills referenced throughout: see § What arrives already decided.
  • references/goal-definitions-and-measurement.md - 针对公司主办开发者活动的每个目标层级定义、每个层级需要的仪表化和合作职能部门、购买者与使用者的细节，以及带有一个正例和一个反例的四字段衡量承诺。
  • references/budget-ownership-and-review.md - 四种所有权模型的细节、共同资助谈判形态和每个资助者在购买什么、每个模型背后的失败模式、离开的所有者必须留下的交接，以及三项审查事项的展开，说明什么可转移什么不可转移。
全文引用的同级skill：见 § 已经决定好的输入。