搜索

PythonReactSEOPDF
GitHub Skill

buyer-journey-mapper

Create persona × journey grids for AEO (Answer Engine Optimization) content strategy. Maps specific questions to buyer stages for AI citation optimization.

majesticlabs-dev☆ 20 仓库 Stars3 仓库 Forks

安装与使用

请根据来源仓库和你的 AI 工具说明安装此技能。先阅读技能文件和依赖要求,再添加到工作流。

查看来源 ↗使用指南

技能说明

Buyer Journey Mapper

Audience: Content strategists optimizing for AI citations Goal: Create question-based content grids mapped to buyer personas and journey stages

The AEO Shift

Move from broad "Ultimate Guides" to specific answers for specific questions.

3×4 Content Grid Template

Create for each product line:

Journey Stage Persona A Persona B Persona C
Awareness "How do I...?" "What is...?" "Why does...?"
Consideration "Best tools for..." "Compare X vs Y" "How to choose..."
Evaluation "Does [Product] do...?" "[Product] vs [Competitor]" "[Product] pricing"
Decision "Can [Product] accomplish...?" "[Product] ROI" "How to implement..."

Persona Specificity

  • ❌ "Marketing Manager"
  • ✅ "Marketing Manager Margaret at a 200-person logistics company managing a 3-person team"

Three-Pronged Query Sourcing

  1. Keyword Data (Ahrefs/SEMrush)

    • Question-based queries
    • "People Also Ask" variations
  2. Social Listening (Reddit, Quora)

    • Community questions
    • Industry forum discussions
  3. Internal Data

    • CRM call transcripts
    • Sales/CS team interviews

Funnel-Stage Query Tagging

Stage Query Pattern Content Type
Awareness "How do I do X?" Educational guides
Consideration "Best tools for ABC" Comparison posts
Evaluation "[Product] features" Feature pages, demos
Decision "Can [Product] accomplish [Task]?" Use cases, ROI calculators

Output

## Buyer Journey Content Grid: [Product/Topic]

### Personas
- Persona A: [specific description]
- Persona B: [specific description]
- Persona C: [specific description]

### Grid
[3×4 table with specific questions]

### Query Sources
- Keyword data: [sources used]
- Social: [communities monitored]
- Internal: [data sources]

### Priority Queue
1. [Highest-opportunity question]
2. [Next priority]
...

使用示例

查看来源仓库中的使用示例与配置说明。

查看来源 ↗

来源历史

收录时间
2026-01-20
索引更新时间
2026-02-04
Branch
main

收录与同步日期属于本站记录。源码修改历史请查看来源仓库。

GitHub ↗

chat Comments (0)

chat_bubble_outline

No comments yet. Be the first to share your thoughts!