AI 提示詞: 電子郵件活動

您的郵件在每個收件匣中都與100多封其他郵件競爭。這些提示詞幫助您打造能被開啟的主旨行、不會讓人厭煩的培養序列,以及真正會被點擊的行動呼籲。已在ChatGPT、Gemini和Claude上測試,讓您知道哪個模型能寫出受眾不會忽視的郵件。

最近測試日期 Feb 15, 2026 · 模型: GPT-4o, Gemini 2.0, Claude 3.5 Sonnet, Grok 2
您想做的事 提示詞 最適合
建立高開信率的郵件主旨行 主旨行機器 ChatGPT
建構新訂閱者的歡迎郵件序列 歡迎序列建立 Claude
重新激活休眠客戶 再互動活動 Gemini
設計長期培養的郵件序列 滴漏式行銷設計 ChatGPT
設計會被閱讀的電子報格式 電子報格式設計 Gemini
規劃郵件的A/B測試 A/B測試規劃器 Claude

主旨行機器

建立高開信率的郵件主旨行

I'm sending an email about [topic/offer/announcement].

Audience: [who they are and what they care about]
Email type: [newsletter/promotional/transactional/re-engagement]
Brand voice: [formal/casual/witty/urgent]
Current open rate: [percentage if known]

Generate:
1. 10 subject lines using curiosity gaps
2. 10 subject lines using specificity and numbers
3. 10 subject lines using personalization placeholders
4. 5 subject lines with matching preview text
5. Rank the top 5 overall with reasoning
6. 3 subject lines to AVOID (common mistakes in [industry])

最適合: CHATGPT

ChatGPT能產生最多樣的主旨行變體,也能有效建立A/B測試的配對。擅長避開垃圾郵件過濾器的措辭。

已測試 Feb 15, 2026

進階技巧

主旨行控制在40個字元以下。考慮手機顯示,較短的主旨行能完整顯示,開信率更高。

歡迎序列建立

建構新訂閱者的歡迎郵件序列

Build a welcome email sequence for new [subscribers/customers/trial users].

Product/service: [what you offer]
Main value proposition: [why they signed up]
Desired end action: [purchase/upgrade/book a call]
Sequence length: [5-7 emails over how many days]

For each email, provide:
1. Send timing (days after signup)
2. Subject line and preview text
3. Full email copy (150-250 words each)
4. The single CTA for each email
5. The psychological principle behind each email's approach
6. Segmentation trigger: what to do if they click vs. don't click

最適合: CLAUDE

Claude能設計最自然流暢的歡迎序列,不強迫推銷但能提供價值的語調出色。

已測試 Feb 15, 2026

進階技巧

設定第一封郵件在註冊後5分鐘內送達。註冊後是互動度最高的時刻。

再互動活動

重新激活休眠客戶

I have [number] subscribers who haven't opened an email in [timeframe].

What they originally signed up for: [lead magnet/offer]
Last email topic they engaged with: [if known]
What's changed since they went cold: [new features, offers, content]

Build a 3-email re-engagement sequence:
1. Email 1: 'We miss you' approach with a compelling reason to return
2. Email 2: Best content/offer roundup of what they've missed
3. Email 3: Final email with clear stay-or-go choice
4. Subject lines for each (optimized for cold subscribers)
5. Win-back incentive recommendations
6. Sunset policy: when to remove unresponsive subscribers and exact criteria

最適合: GEMINI

ChatGPT能建立最有效的再互動郵件,不顯得強迫地傳達「歡迎回來」的訊息。

已測試 Feb 15, 2026

進階技巧

在再互動之前,先分析為什麼流失。在不了解原因的情況下重複相同的做法沒有效果。

滴漏式行銷設計

設計長期培養的郵件序列

I need an automated drip campaign for [purpose: lead nurture / onboarding / upsell / event].

Trigger event: [what starts the sequence]
Audience segment: [who enters this campaign]
Campaign goal: [desired conversion]
Timeline: [days/weeks the campaign runs]
Existing content assets: [blog posts, case studies, webinars to reference]

Design the complete campaign:
1. Campaign flowchart with decision branches (if opened/clicked vs. didn't)
2. Email copy for each node (subject, body, CTA)
3. Wait times between emails with reasoning
4. Branch logic: what happens at each decision point
5. Exit conditions: when someone leaves the campaign
6. KPIs to track at each stage and benchmarks to aim for

最適合: CHATGPT

Gemini能設計最有邏輯的滴漏式序列,各郵件之間的時間間隔和內容進展出色。

已測試 Feb 15, 2026

進階技巧

郵件之間的最佳間隔因受眾而異。B2B一般3到5天,B2C一般1到3天,但透過測試來最佳化。

電子報格式設計

設計會被閱讀的電子報格式

I want to launch a [frequency] newsletter for [audience].

Topics I'll cover: [3-5 content areas]
Tone: [educational/entertaining/curated/opinionated]
Time I can spend per issue: [hours]
Monetization plan: [ads/sponsorships/premium/lead gen/none yet]

Design my newsletter:
1. A name and tagline (5 options with reasoning)
2. A repeatable section structure (e.g. intro, main story, quick hits, CTA)
3. Word count targets for each section
4. A content sourcing system: where to find material each week
5. A sample first issue with full copy
6. Growth strategy for the first 1,000 subscribers

最適合: GEMINI

ChatGPT能設計最易讀的電子報版面,段落分隔和視覺層次清晰。

已測試 Feb 15, 2026

進階技巧

電子報的長度目標是500字以下。太長的電子報即使被開啟也不會被讀完。

A/B測試規劃器

規劃郵件的A/B測試

I want to A/B test my emails to improve [open rate/click rate/conversion rate].

Current metrics: [open rate, click rate, list size]
Email platform: [Mailchimp/ConvertKit/ActiveCampaign/etc.]
Send frequency: [how often you email]
Biggest unknown: [what you're most unsure about]

Build a 4-week testing plan:
1. Week 1 test: what to test, hypothesis, and how to measure
2. Week 2 test: building on week 1 results
3. Week 3 test: a more advanced element to test
4. Week 4 test: final optimization
5. Minimum sample size needed per test for statistical significance
6. A testing rules document: when to call a winner and when to keep testing

最適合: CLAUDE

Gemini能建立最具統計學基礎的A/B測試計劃,樣本大小的計算和測試期間的建議準確。

已測試 Feb 15, 2026

進階技巧

一次只測試一個變數。同時改變主旨行和內文,就無法知道是哪個影響了效果。

模型比較

基於實際測試結果 — 非假設推測。 查看測試方法

G

Gemini

Best for re-engagement campaigns and newsletter design. Creates sustainable email formats with realistic production timelines. Less effective at emotional or story-driven copy.

結果來源: Gemini 2.0 Flash · 已測試 Feb 15, 2026
C

ChatGPT

Best for subject lines, drip campaign logic, and platform-specific copy. Generates the most natural-sounding email copy but can over-optimize for clicks at the expense of authenticity.

結果來源: GPT-4o · 已測試 Feb 15, 2026
C

Claude

Best for welcome sequences and A/B test design. Builds psychologically sophisticated email journeys and catches statistical errors in testing plans. Sometimes writes emails that are too long.

結果來源: Claude 3.5 Sonnet · 已測試 Feb 15, 2026
G

Grok

Generates the punchiest subject lines and writes emails with a distinctive, scroll-stopping voice. Strong at creating urgency without sounding desperate. Less effective at building long-form nurturing sequences that require patience and gradual trust-building.

結果來源: Grok 2 · 已測試 Feb 15, 2026

在 NailedIt 中試試

將上方的提示詞貼到 NailedIt,並排比較各模型的回應。

進階技巧

1

Write the PS line first. The PS is the second most-read part of any email after the subject line. Write it before the body and put your most important message there. Many people scroll straight to the bottom.

2

One email, one ask. Every email with multiple CTAs converts worse than one with a single clear action. If you have three things to say, send three emails. AI will try to stuff everything in — tell it to pick one CTA.

3

Send time matters less than you think. The difference between the 'best' and 'worst' send time is usually 2-3% open rate. Spend that optimization energy on better subject lines instead — they swing open rates by 20%+.