Files
nexus/wiki/concepts/Lost-Prompt-Analysis.md
2026-04-26 08:02:48 +08:00

51 lines
1.9 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
---
title: "Lost Prompt Analysis"
type: concept
tags: ["AEO", "GEO", "analysis", "competitive-intelligence"]
last_updated: 2026-04-26
---
## Definition
Lost Prompt Analysis丢失提示分析是 AI Citation Audit 的核心分析方法之一——系统性地识别品牌应该出现在 AI 答案中但竞争对手胜出的查询场景。通过分析"为什么会输",推导出内容修复方向。
## Method
1. **Query Set Generation**:生成 20-40 个目标受众会在 AI 平台输入的查询
- 类型:推荐类、对比类、教程类、最佳选择类
- 格式:"Best X for Y"、"X vs Y"、"How to choose X"、"Recommend a X"
2. **Multi-Platform Query**:在 ChatGPT、Claude、Gemini、Perplexity 四个平台分别运行完整查询集
3. **Result Recording**:记录每次查询中:
- 哪个品牌被引用
- 引用位置(开头/中间/结尾)
- 引用上下文(作为什么类型的来源被引用)
4. **Gap Identification**:标记品牌缺席但竞品出现的查询 → **Lost Prompts**
5. **Root Cause Analysis**:对每个 Lost Prompt 分析竞品胜出的原因:
- 内容结构FAQ 页 vs 博客文章)
- Schema markup有无 FAQPage/Product schema
- 实体信号(品牌一致性、知识图谱覆盖)
- 内容格式(对比表格 vs 段落叙述)
## Output Format
```markdown
| Prompt | Platform | Who Gets Cited | Why They Win | Fix Priority |
|--------|---------|---------------|--------------|-------------|
| "Best X for Y" | All 4 | Competitor A | Comparison page with structured data | P1 |
| "How to choose X" | ChatGPT, Gemini | Competitor B | FAQ matching query pattern | P1 |
```
## Related Concepts
- [[Fix Pack]]Lost Prompt Analysis 的产出驱动 Fix Pack 的生成
- [[Citation Rate]]Lost Prompt 数量直接影响整体 Citation Rate
- [[Platform-Specific Patterns]]:不同平台 Loss Prompt 的原因可能不同
## Sources
- [[AI Citation Strategist]]