R「声誉乘数」
正式名称 R「声誉乘数」(Reputation Multiplier)。由先赋、受众、历史评价、预期、贡献意愿五因子构成,连乘后除以 1,000,对照锚是 R=1(不放大也不抑制);使用中因子不低于 1。贡献意愿指受众愿不愿意为你出力,不是主体自己对外付出。不是百分制考卷。兼容锚点 #r。官网免费线上测试目前只做这一项。
以下为中文公开说明,供首页、理论与工具页引用。不包含合同附件、内部 SOP 或尚未对外锁定的阈值数字。英文原工具说明与可运行工作表保留在本页后部。使用须经授权,详见法律声明。
正式名称 R「声誉乘数」(Reputation Multiplier)。由先赋、受众、历史评价、预期、贡献意愿五因子构成,连乘后除以 1,000,对照锚是 R=1(不放大也不抑制);使用中因子不低于 1。贡献意愿指受众愿不愿意为你出力,不是主体自己对外付出。不是百分制考卷。兼容锚点 #r。官网免费线上测试目前只做这一项。
EIC(期望激励资本)= V × E × R,用于盘点现状并形成可对照的判断,而不是给出一个可对外夸示的单一排名。EIC、RIC、AHP 同属咨询执行工具:不进入线上 R「声誉乘数」,也不互相取代。不在免费测评页做在线追踪。
RIC 是 Reach / Interpretation / Consensus,即触达、解读、共识。三块分看,不合成一个总分,也不与 R「声誉乘数」混加或互相取代。用于判断信息有没有到达、有没有被正确理解、有没有形成可核验的共识。
声誉钢印指可核验、可重复调用的声誉印记,而不是口头承诺或一次性曝光。它强调留下能被第三方再次找到的记录,而不是“说过就算数”。
面向持续防御与复测的方法组合:把可能出事的点逐条列出、定级、配预案与责任人,目的是把临时应对变成可运转的制度。
判断搜索与生成式引擎中,主体被如何描述与调用,并给出可执行的优化路径。我们不承诺平台改口。开工前须具备基本的公开数字信息存量。
在可控情景中预演决策速度、口径一致性与角色分工。它测的是团队在压力下会不会各说各话,而不是再听一遍课。
整理可核验的公开信源与调用关系,为判断和回应提供依据。它不是监控软件的替代品,也不作为独立监测产品销售。
层次分析法(Analytic Hierarchy Process),用于在多因素之间做权重分配与优先级判断,避免只凭感觉排顺序。AHP 是咨询执行工具,不进入线上 R,也不与 EIC、RIC 互相取代。
按谣言的生成与传播机制分类,便于对症处置,而不是一律用同一套公关回应。公开页只说明分类用途,不展开内部操作细则。
用五因子中的维度做对照分布,帮助看出结构而不是只看一个总分。F5(贡献意愿)不作轴,须另行标注。
把议题、责任与动作拆成相互独立、尽量穷尽的对照表,避免遗漏和重叠,便于分工和验收。
因信任不足而多付出的说服、担保或交易成本。规则待锁定,本页只作方法说明,不给出对外计价公式或阈值。
在有限、可核验的样本上做前后或组别对照,避免用不可比的大数代替判断。
用于判断公开数字信息存量是否足以开展抓取或诊断。不够则停抓取、改走自测。阈值数字尚未对外锁定。
按英文书口径,日常盘点通常覆盖:顾客、投资者与资本提供方、员工与人才、监管者、媒体与信息中介、合作伙伴、社区与公民社会。
意见领袖、狂热分子、背书人、缄默者、八卦者、乌合之众、叛逆者。仅作为危机公关识别工具展示,不进入在线测试,也不作为分享标签。
Every tool below is described in the appendix of the book. Three of them also run in the browser, further down this page — no account, no storage, nothing sent anywhere.
One row for each segment: customers; investors and capital providers; employees and talent; regulators; media and information intermediaries; partners; communities and civil society. Three columns for reach, interpretation, and consensus, and two more for the data source and the date the number was taken.
One page per cycle, three blocks, never one total. Each block carries a definition, a data source, a sample, and a statement of uncertainty. Record the direction and size of change, not only the level.
Score each factor from 1 to 10 and write the reason beside the number. Then re-weight the factors for each audience and for each stage of the company's development, and re-run the scorecard on a fixed calendar.
Qualified promises kept, divided by promises made. Read per audience and per year, it is the most direct single measure of the expectation factor, and it moves before survey scores do.
To measure contribution willingness without asking a leading question, quote different price points to comparable customer cohorts — five to thirty percent apart — over thirty days, and watch behaviour rather than stated intent.
Take ten transactions from the past year. For each, record what you spent in time, money, deposits, legal review, and waiting to be believed; then estimate what a counterparty with a strong reputation would have spent for the same transaction.
Tag every monitored mention by issue and by sentiment, then read the matrix rather than the total.
Realised return divided by promised return, read against the resources invested.
Plot a live crisis on two axes — severity of harm and speed of spread — to decide whether to answer with facts, with action, or with both.
Tell it fast, tell it all, tell your own tale — but in sequence rather than at once. The crisis team holds the whole picture, one spokesperson does the talking, and the message leaves from a single point.
Type your own numbers. Nothing you enter leaves this page: there is no account, no cookie, and no request to a server. Reload and it is gone.
Score each factor from 1 to 10. Be honest with the weakest one; it is the only one worth acting on first.
Qualified promises kept, divided by promises made. Read it per audience and per year — never as a company-wide average.
For one transaction: what you spent to be believed, against what a counterparty with a strong reputation would have spent for the same thing.