← 行銷專欄 Marketing · 行銷學習地圖 W1-2

行銷基礎與 2025-26 趨勢

Marketing Foundations & 2025-26 Trends

漏斗思維的百年演進、內容與廣告的取捨，以及 AI 時代行銷的三大變局。

A century of funnel thinking, the content-vs-ads tradeoff, and three seismic shifts in AI-era marketing.

🎯 學完你能做到 🎯 Learning Goals ① 為任何產品選對框架（AIDA / AARRR / Flywheel）② 解釋 AEO/GEO 與 zero-click 的成因與對策 ③ 用 JTBD、StoryBrand、Hook Model 各分析一個真實產品 ① Pick the right framework (AIDA / AARRR / Flywheel) for any product ② Explain AEO/GEO and zero-click, with countermeasures ③ Analyze a real product with JTBD, StoryBrand, and the Hook Model
🔻
漏斗演進三部曲：AIDA → AARRR → Flywheel
Funnel Evolution: AIDA → AARRR → Flywheel
1898 → 2007 → 2018
Week 1必讀 Core
▾
🗺️ 三者是巢狀關係，不是取代關係Nested, Not Replacing
AIDA
單次轉換 conversion copy
⊂
AARRR
指標體系 metrics system
⊂
Flywheel
整體成長觀 growth model

最常見的錯誤是把三者當「新淘汰舊」。實際上 AIDA 管一頁文案的說服結構、AARRR 管全公司的量測體系、Flywheel 管顧客如何成為成長動力——三層同時存在。

The most common mistake is treating these as "new replaces old." In reality AIDA governs persuasion within a single page, AARRR governs company-wide measurement, and the Flywheel governs how customers power growth — all three layers coexist.

1️⃣ AIDA（1898, E. St. Elmo Lewis）
  • Awareness → Interest → Desire → Action，最古老的線性漏斗
  • 現代定位：轉換文案層面仍然有效——sales page、landing page、廣告腳本的段落結構
  • Modern role: still valid for conversion copy — sales pages, landing pages, ad scripts
  • 侷限：只走到成交為止，不處理留存與推薦；真實消費者會跳階、回流
  • Limit: stops at the sale; ignores retention & referral; real consumers skip and loop back
誤區：宣稱「AIDA 過時該淘汰」。錯的是拿它當整體成長模型，不是框架本身。Pitfall: declaring "AIDA is dead." What's wrong is using it as a growth model, not the framework itself.
2️⃣ AARRR / Pirate Metrics（2007, Dave McClure）
階段 Stage問題Question範例 KPIExample KPI
Acquisition 獲客使用者從哪來？Where do users come from?Sessions by source/mediumSessions by source
Activation 啟用第一次就體驗到價值？First-run value (aha moment)?完成關鍵動作率Key action completion
Retention 留存會回來嗎？Do they return?D7 / D30 retentionD7 / D30 retention
Revenue 營收願意付錢嗎？Do they pay?ARPU, conversion rateARPU
Referral 推薦會帶朋友來嗎？Do they refer?NPS, referral sessionsNPS
  • 核心價值：每階段對應可量測 KPI，讓團隊脫離 vanity metrics（虛榮指標）
  • Core value: every stage maps to a measurable KPI, escaping vanity metrics
  • 變體 RARRA：把 Retention 排最前——反映「留存比獲客重要」的現代共識
  • Variant RARRA: Retention first — reflecting the modern consensus that retention beats acquisition
誤區：一次優化全部五階段。正確做法是找出漏水最嚴重的一段（one metric that matters）集中火力。Pitfall: optimizing all five stages at once. Find the leakiest stage (one metric that matters) and focus.
3️⃣ Flywheel（2018, HubSpot）
  • Attract → Engage → Delight 循環，顧客位於圓心 loop, with the customer at the center
  • 核心論點：漏斗把顧客當「產出物」（掉出底部就結束）；飛輪把顧客當「動力來源」——口碑、評論、推薦帶動下一輪成長
  • Core thesis: funnels treat customers as output; the flywheel treats them as the force — word-of-mouth, reviews, referrals power the next cycle
  • 兩個操作變數：加「力」force（投資顧客成功）、減「摩擦」friction（移除購買與使用障礙）
  • Two levers: add force (invest in customer success), remove friction (eliminate obstacles)
  • 最適用：訂閱制、社群驅動、口碑效應強的產品
  • Best fit: subscriptions, community-driven, word-of-mouth products
誤區：以為飛輪取代漏斗所以不用管轉換率——飛輪內部每個階段仍然有轉換漏斗。Pitfall: assuming the flywheel replaces funnels — every flywheel stage still contains conversion funnels.
⚖️
內容行銷 vs 付費廣告
Content Marketing vs Paid Ads
資產 vs 租金
Asset vs Rent
Week 1策略 Strategy
▾
🔑 核心觀念Core Concepts
內容行銷 ContentContent付費廣告 Paid AdsPaid Ads
本質Nature資產（複利累積）Asset (compounds)租金（停付歸零）Rent (stops when paused)
見效Speed6-12 個月見複利6-12 months to compound立即Immediate
適合Best for長期信任、教育型產品、SEO 資產Trust, education, SEO assets限時活動、放大已驗證漏斗、受眾測試Launches, scaling proven funnels, testing
  • 最佳實務是混合：廣告導流到高品質內容 → 內容養信任 → 轉換 → 顧客帶來 organic referral
  • Best practice is hybrid: ads drive traffic to quality content → content builds trust → converts → customers refer organically
  • 先用有機內容驗證訊息（哪個 hook 有共鳴），再用廣告放大——別花錢推沒驗證過的訊息
  • Validate messaging organically first, then amplify with ads — never pay to push unvalidated messages
🧭
三個必學框架：JTBD、StoryBrand、Hook Model
3 Key Frameworks: JTBD, StoryBrand, Hook Model
需求洞察 · 品牌敘事 · 習慣設計
Demand Insight · Brand Narrative · Habit Design
Week 2必讀 Core
▾
🥤 Jobs-to-be-Done（Clayton Christensen）
  • 核心問句：「顧客雇用你的產品來完成什麼任務（job）？」
  • Core question: "What job does the customer hire your product to do?"
  • 奶昔案例：McDonald's 奶昔多在早晨售出——顧客雇用它「陪伴無聊通勤 + 單手可食 + 撐到中午」。競品不是其他甜點，是貝果、香蕉、無聊本身。做更濃稠後銷量成長
  • Milkshake case: morning milkshakes were hired to "accompany a boring commute + one-handed + last till noon." Competitors weren't desserts — they were bagels, bananas, and boredom itself
  • Job 三維度：functional（功能）、emotional（情緒）、social（社交）
  • Three job dimensions: functional, emotional, social
誤區：做成人口統計 persona——JTBD 刻意去人口統計化，同一個人在不同情境雇用不同產品。Pitfall: turning JTBD into demographic personas — JTBD is deliberately demographic-free; the same person hires different products in different situations.
📖 StoryBrand SB7（Donald Miller）
Character
顧客=主角
→
Problem
→
Guide
品牌=嚮導
→
Plan
→
CTA
→
Failure / Success
  • 最重要的翻轉：顧客是英雄、品牌是嚮導（Yoda 之於 Luke）——多數品牌失敗因為把自己當英雄
  • The key inversion: the customer is the hero, the brand is the guide (Yoda to Luke) — most brands fail by casting themselves as the hero
  • 名言："If you confuse, you lose"——顧客買講得最清楚的，不是講得最聰明的
  • "If you confuse, you lose" — customers buy from whoever is clearest, not cleverest
  • 適用：官網首頁訊息、品牌敘事、elevator pitch、email 序列
  • Use for: homepage messaging, brand narrative, elevator pitch, email sequences
誤區：Problem 只寫外部問題——SB7 要求同時處理 internal（挫折感）與 philosophical（這不應該發生）層次，內部問題才是購買動機。Pitfall: only writing external problems — SB7 requires internal (frustration) and philosophical layers too; internal problems drive purchases.
🪝 Hook Model（Nir Eyal）
Trigger
外部→內部觸發
→
Action
最低阻力行為
→
Variable Reward
變動獎勵
→
Investment
用戶投入
  • 變動獎勵三類：Tribe（社交認同）、Hunt（資源獵取，如無限滾動）、Self（自我掌握感）
  • Three variable rewards: Tribe (social), Hunt (resources, e.g. infinite scroll), Self (mastery)
  • 行銷應用：內容系列化（養成回訪習慣）、email/推播設計、產品 onboarding
  • Marketing use: serialized content (habit-forming return visits), email/push design, onboarding
  • 附帶倫理：Eyal 本人提出 Manipulation Matrix 自我檢驗——負責任的習慣設計
  • Ethics: Eyal's own Manipulation Matrix for responsible habit design
誤區：獎勵做成固定（可預測的獎勵不形成習慣）；跳過 Investment 步驟（用戶無沉澱、隨時流失）。Pitfall: fixed rewards (predictable rewards don't form habits); skipping Investment (no user deposit, easy churn).
✏️
實作練習
Practice Exercises
用自己的專案練，不是紙上談兵
Practice on your real projects
動手做 Hands-on
▾
練習 1（W1）：AARRR 健檢 Exercise 1 (W1): AARRR Audit 用 AARRR 幫你的產品（如 telaaurealab.com）畫出五階段現況與 KPI：Acquisition = GA4 流量來源、Activation = 使用者第一次生成圖案、Revenue = 付款完成……找出漏水最嚴重的一段，寫下一個改善假設。 Map your product's five AARRR stages with KPIs (Acquisition = GA4 sources, Activation = first pattern generated, Revenue = checkout…). Identify the leakiest stage and write one improvement hypothesis.
練習 2（W2）：JTBD 陳述 Exercise 2 (W2): JTBD Statement 為你的產品寫一份 JTBD：誰、在什麼情境、雇用它完成什麼 job？functional / emotional / social 各是什麼？（例：手作愛好者雇用捕夢網生成器，在「想送獨一無二的禮物」情境下，完成「快速做出有個人意義的設計」——emotional job 是心意的表達。） Write a JTBD statement: who, in what situation, hires it for what job? What are the functional / emotional / social dimensions?
練習 3（W2）：SB7 重寫首頁 Exercise 3 (W2): SB7 Homepage Rewrite 用 StoryBrand 七元素重寫你網站首頁的 hero 區塊文案：顧客是英雄（不是你的產品）、指出 internal problem、你是給計畫的嚮導、一個清楚的 CTA。 Rewrite your homepage hero copy with SB7: customer as hero, name the internal problem, position yourself as the guide with a plan, one clear CTA.
📖

術語速查Glossary