pain-point-marketing-loop

📁 atxinsky/skills 📅 2 days ago
1
总安装量
1
周安装量
#46759
全站排名
安装命令
npx skills add https://github.com/atxinsky/skills --skill pain-point-marketing-loop

Agent 安装分布

amp 1
opencode 1
kimi-cli 1
codex 1
github-copilot 1
claude-code 1

Skill 文档

Pain Point Marketing Loop

Overview

Systematic methodology for identifying user pain points through social media analysis, creating targeted solutions, and building conversion funnels with minimal cost. Core principle: Find real pain → Create authentic solutions → Convert through value exchange.

When to Use

  • Starting content marketing or user acquisition campaigns
  • Need to understand target audience pain points deeply
  • Want to build conversion funnels with low CAC (Customer Acquisition Cost)
  • Have limited marketing budget but good content creation skills
  • Targeting specific demographics with known pain points

When NOT to use:

  • High-budget campaigns where paid ads are more efficient
  • B2B enterprise sales requiring personal relationships
  • Products with no clear user pain points
  • When you lack content creation or social media analysis skills

Core Pattern

Step 1: Pain Point Mining (Data Collection)

# Search social platforms for pain points
keywords = ["35岁", "焦虑", "工作累回家更累", "中年危机"]
platforms = ["小红书", "知乎", "抖音", "B站"]

# Collect high-engagement comments (likes > threshold)
pain_points = collect_comments(keywords, min_likes=10, count=100)

Step 2: AI Analysis (Pattern Recognition)

# Feed to AI for pain point extraction
prompt = f"分析这些用户评论,找出最痛的三点:\n{pain_points}"
top_pain_points = ai_analyze(prompt, model="claude-3")

Step 3: Content Creation (Solution Development)

# Generate comprehensive solutions
for pain in top_pain_points:
    report = ai_generate_solution_report(pain, format="PDF")
    mindmap = ai_generate_mindmap(pain, format="XMind")

Step 4: Traffic Conversion (Value Exchange)

# Reply to original comments with personalized solutions
for comment in high_engagement_comments:
    personalized_reply = customize_solution(comment, user_profile)
    reply_to_comment(comment, personalized_reply)

Step 5: User Cultivation (Trust Building)

# Free value exchange builds trust
if user_requests_resource:
    send_free_resource(user)
    add_to_nurture_list(user)

Step 6: Soft Launch (Organic Growth)

# 2-3 days later, soft promotion
post_friends_circle(soft_ad_copy, target_audience)

Implementation

Tools Required

  • Social media monitoring tools (manual or automated)
  • AI content generation (Claude, GPT, etc.)
  • PDF/Mindmap creation tools
  • Social media management platform

Success Metrics

  • Conversion Rate: 5-10% (industry good)
  • CAC: <$5 per customer
  • Content Quality: 80%+ engagement rate
  • User Lifetime Value: >3x CAC

Common Pain Points by Demographics

Demographic Common Pain Points Solution Type
35岁职场人 工作疲惫、家庭压力、精力不足 精力恢复、时间管理
宝妈 育儿焦虑、自我价值、社交孤立 亲子关系、自我成长
创业者 资金压力、决策焦虑、人脉拓展 商业思维、资源整合
学生党 学习压力、职业迷茫、人际关系 学习方法、生涯规划

Quick Reference

Step-by-Step Checklist

  • 收集100+高赞评论
  • AI提取top 3痛点
  • 生成解决方案报告
  • 个性化回复导流
  • 免费提供价值
  • 软文朋友圈推广
  • 追踪转化效果

Content Templates

痛点提取Prompt:

分析以下用户评论,找出最普遍、最深刻的三个痛点:
[粘贴评论内容]

请按重要性排序,给出具体表现和影响。

解决方案报告Prompt:

基于这个痛点:[痛点描述]
创建一个详细的解决方案报告,包含:
1. 痛点分析
2. 科学解决方案
3. 实施步骤
4. 预期效果
5. 注意事项

Conversion Scripts

评论回复模板: “看到你的分享,很有共鸣。我也经历过类似的情况。最近整理了一份关于[解决方案]的资料,帮我改善了很多。[个性化建议]。如果需要,可以私信我免费领取。”

Common Mistakes

❌ Too Generic Content

Problem: “我觉得你说的对,我也这样” – 被平台限流 Fix: 个性化回复,提具体建议或共鸣点

❌ Hard Selling Too Early

Problem: 直接推销产品,用户反感 Fix: 先提供价值,建立信任,2-3天后软推广

❌ Insufficient Data Sampling

Problem: 只收集20条评论,痛点不准 Fix: 最少100条,覆盖不同时间段和平台

❌ No Follow-up System

Problem: 发了资料就没了,用户流失 Fix: 建立用户群或邮件列表,持续提供价值

Real-World Impact

Case Study – 精力恢复报告:

  • 投入: 2小时内容收集 + 1小时AI创作 + 3小时回复
  • 产出: 50个精准用户,15个转化,平均客单价¥199
  • ROI: 15x (投资回收15倍)

Key Success Factors:

  • 真实痛点 > 营销话术
  • 免费价值 > 直接销售
  • 持续跟进 > 一锤子买卖

Related Skills

  • research: 用于深入分析痛点数据
  • frontend-design: 创建解决方案报告页面
  • notion-save: 保存用户数据和跟进记录