From marketing-strategy
Use when the user wants to apply the S-O-R model, Stimulus-Organism-Response, 消費者行為分析, 行銷心理, 刺激-有機體-反應, or to turn 廣告、CRM、落地頁、電商、會員、回購、KPI、A/B 測試 ideas into a structured marketing strategy, psychology map, measurement plan, or experiment brief.
How this skill is triggered — by the user, by Claude, or both
Slash command
/marketing-strategy:sor-marketing-strategyThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
使用這個技能,把外部刺激(Stimulus)、消費者內在心理歷程(Organism)、外顯行為反應(Response)串成可執行的行銷策略,而不是只停留在理論解說。
使用這個技能,把外部刺激(Stimulus)、消費者內在心理歷程(Organism)、外顯行為反應(Response)串成可執行的行銷策略,而不是只停留在理論解說。
先做 S-O-R 分析,再輸出策略、KPI、實驗與風險控管。資料不足時,先補資料,不硬推結論。
Use:
S-O-R、Stimulus-Organism-Response、刺激-有機體-反應。Do not use:
Handoff:
message_strategy_brief,再建議轉交 $copywriting。experiment_plan 骨架,再建議轉交 $ab-test-setup。kpi_ladder 與事件需求,再建議轉交 $analytics-tracking。必要欄位:
brand_or_offer: stringtarget_audience: stringbusiness_goal: stringchannel_context: string[]desired_response: attention|engagement|conversion|repurchase|advocacy可選欄位:
funnel_stage: awareness|consideration|conversion|retention|reactivationcurrent_stimuli: string[]proof_assets: string[]constraints: string[]timeline: stringbudget: stringavailable_metrics: string[]market_context: stringcompetition_context: stringcustomer_insights: string[]risk_notes: string[]缺少必要欄位時,不得直接產出完整策略,先回傳 MissingDataOutput:
{
"missing_fields": [],
"why_needed": {},
"questions_to_user": [],
"next_step_rule": "補齊必要欄位後再產出完整 S-O-R 策略"
}
最低標準:
固定角色:
輸出必須具備:
Stimulus -> Organism -> Response禁止事項:
先辨識現有刺激與缺口,再補最有機會影響目標行為的刺激。
Stimulus 五類:
sensory:視覺、版面、色彩、動線、門市氛圍、素材節奏informational:價值主張、產品資訊、價格、比較、保證、FAQsocial:評論、案例、社群互動、UGC、專家或同儕背書technical:介面回饋、推播、推薦、個人化、速度、流程摩擦situational:時間點、檔期、場景、裝置、使用情境、回購時機每次輸出都要回答:
把消費者內在歷程拆成以下五類:
cognitive:是否看懂、是否容易比較、是否降低認知負荷emotional:是否感到被理解、被重視、安心、興奮、好奇attitude_brand_affinity:是否形成正向態度、品牌好感與「這家店很貼心」的判斷trust_risk:是否相信品牌、是否覺得風險可控value_appraisal:是否認為值得、相關、划算、符合自我需求intrusiveness_friction:是否覺得被打擾、被監視、被硬推、被迫行動判讀原則:
Response 依商業成熟度拆成五級:
attention:看到、停留、記住、點開engagement:互動、收藏、加入購物車、回覆、深入瀏覽conversion:註冊、下單、預約、留單、付款repurchase:回購、續約、升級、再次到訪advocacy:推薦、評論、分享、UGC、介紹他人輸出時要清楚標示:
desired_response,再倒推需要的 organism,再設計 stimulus。repurchase 或 advocacy,必須補 retention 維度,不可只看單次轉換。完整輸出必須符合 StrategyOutput:
{
"executive_summary": [],
"sor_map": {
"stimulus": [],
"organism": [],
"response": []
},
"stimulus_design_measures": [],
"organism_influence_measures": [],
"response_measurement_measures": [],
"strategy_actions": [],
"channel_plan": [],
"kpi_ladder": [],
"experiment_plan": [],
"optimization_loop": {
"response_data_to_review": [],
"low_efficiency_stimuli": [],
"tuning_actions": [],
"next_test": ""
},
"risks_and_ethics": [],
"next_30_day_actions": []
}
欄位要求:
executive_summary:2-5 點關鍵判斷sor_map:至少各列出 3 個高影響因子,並串成因果鏈stimulus_design_measures:對應草稿的「Stimulus 設計方式」,至少列 3 項具體措施organism_influence_measures:對應草稿的「Organism 影響方式」,至少列 3 項具體措施response_measurement_measures:對應草稿的「Response 測量方式」,至少列 3 項具體措施strategy_actions:每項行動包含 objective、stimulus_change、expected_organism_shift、target_responsechannel_plan:至少標出 channel、message_role、timingkpi_ladder:每個目標 response 都要有 leading 與 lagging 指標experiment_plan:列出優先測試假設、control/variant 概念、風險與最小可行測試optimization_loop:固定包含 response_data_to_review、low_efficiency_stimuli、tuning_actions、next_testrisks_and_ethics:至少覆蓋 trust、intrusiveness、measurement gapsnext_30_day_actions:拆成週或階段若使用者其實要文案方向,再多輸出:
{
"message_strategy_brief": {
"core_tension": "",
"message_angles": [],
"proof_points": [],
"cta_directions": []
}
}
caution。Assumption 與 Validation Needed。請用 $sor-marketing-strategy 分析這個落地頁的刺激、心理機制與轉換問題。我們想提升會員回購,請用 S-O-R 模型設計 CRM 與 KPI ladder。請用 S-O-R 拆解這個品牌活動為什麼有聲量卻沒轉單。請用 $sor-marketing-strategy 先做策略與 message brief,文案先不要直接成稿。npx claudepluginhub timlai666/skills --plugin marketing-strategyGuides collaborative design exploration before implementation: explores context, asks clarifying questions, proposes approaches, and writes a design doc for user approval.
Creates structured, bite-sized implementation plans from specs or requirements before writing code. Useful for breaking down multi-step tasks into testable steps with file structure and task boundaries.
Synthesizes the current conversation into a structured spec (PRD) and publishes it to the project issue tracker with a ready-for-agent label, without interviewing the user.