사용 방법
- 설치 명령을 복사하고 자신의 에이전트 환경에 설치하기 전에 소스 패키지를 확인하세요.
- 필요한 도구, 인증 정보, 지원 환경은 원본 문서에서 확인하세요.
- 입력과 원하는 산출물을 설명하고 출처와 결과를 대조하세요.
npx skills add https://github.com/apidojo-io/apidojo-skills --skill extracting-google-paa-questions-for-seo시작 전 확인
- 설치에는 모델 구독, API 한도, 파일 접근 권한이 포함되지 않습니다.
- 별점 수와 설치 수는 활동 지표이며 독립적인 품질·보안 검증이 아닙니다.
원본 저장소의 제목, 예시, 코드, 표, 링크, 이미지를 유지해 표시합니다.
Extracting Google PAA Questions for SEO
Pulls People Also Ask (PAA) questions from Google SERPs for a target keyword. PAA questions are Google-validated signals of what real users want to know — use them as H2/H3 headings or FAQ sections in content.
Prerequisites
APIFY_TOKENenvironment variable set- Optional: Apify MCP server installed
Inputs
| Parameter | Type | Required | Default | Notes |
|---|---|---|---|---|
startUrls | array | Optional | [] | Google search URLs |
searchTerms | array | Optional | [] | Keywords to search on Google |
countryCode | string | Optional | US | Country for Google search (e.g. US, GB, TR) |
languageCode | string | Optional | — | Language for results (e.g. en) |
maxItems | number | Optional | Unlimited | Maximum results to return across all queries |
maxPagesPerQuery | integer | Optional | 1 | Maximum result pages per query |
mobileResults | boolean | Optional | false | Fetch mobile SERP layout |
customMapFunction | string | Optional | — | JavaScript function to transform each output object |
Workflow
Progress:
- [ ] Step 1: Run google-search-scraper for keyword and variants
- [ ] Step 2: Extract PAA questions from results
- [ ] Step 3: Classify questions by search intent
- [ ] Step 4: Score content opportunity per question
- [ ] Step 5: Deliver SEO content briefStep 1: Run google-search-scraper
Recommended — run_actor.js (handles waiting, output, and file saving automatically):
# Quick answer (prints table to chat)
node scripts/run_actor.js \
--actor "apidojo~google-search-scraper" \
--input '{"param": "value"}'
# Save as CSV
node scripts/run_actor.js \
--actor "apidojo~google-search-scraper" \
--input '{"param": "value"}' \
--output YYYY-MM-DD_results.csv --format csv
# Save as JSON
node scripts/run_actor.js \
--actor "apidojo~google-search-scraper" \
--input '{"param": "value"}' \
--output YYYY-MM-DD_results.json --format jsonAPIFY_TOKENmust be set in environment or.envfile.
If Apify MCP is available:
Tool: apify:run-actor
Actor: "apidojo~google-search-scraper"
Input:
{
"queries": ["[KEYWORD]", "best [KEYWORD]", "how to [KEYWORD]", "[KEYWORD] for beginners"],
"maxPagesPerQuery": 3,
"countryCode": "US",
"languageCode": "en",
"includePeopleAlsoAsk": true
}REST API fallback:
curl -X POST "https://api.apify.com/v2/acts/apidojo~google-search-scraper/runs?token=$APIFY_TOKEN" -H "Content-Type: application/json" -d '{
"queries": ["email marketing tools", "best email marketing tools", "how to email marketing"],
"maxPagesPerQuery": 3,
"countryCode": "US",
"includePeopleAlsoAsk": true
}'Step 2: Classify PAA Questions by Intent
intent:
INFORMATIONAL = "what is", "how does", "what does", "explain"
COMPARATIVE = "vs", "or", "better", "difference between"
TRANSACTIONAL = "how to buy", "where to get", "best price", "free"
NAVIGATIONAL = "[brand] how to", "login", "sign up"
TROUBLESHOOTING = "not working", "error", "fix", "issue"Step 3: Score Content Opportunity
paa_score = (question_appears_across_multiple_keywords ? 1 : 0.5) * 0.35
+ (intent == INFORMATIONAL ? 1 : 0.7) * 0.30
+ (current_SERP_has_no_featured_snippet ? 1 : 0.4) * 0.35Featured snippet opportunity: if PAA answer shown is > 200 words or from a weak domain → high opportunity to claim.
Step 4: Edge Cases
- PAA questions not returned: Some queries return no PAA boxes; try more question-form queries ("how to [keyword]", "what is [keyword]")
- Duplicate questions across keywords: Deduplicate by question text similarity (≥ 80% overlap = same question); count as one, note it appeared for [N] keywords
- Questions are too broad: Flag as
BROAD_QUESTION— better suited for a pillar page or FAQ section, not a standalone post - Questions are brand-specific competitors: Include with
COMPETITIVE_INTELLIGENCEflag — these tell you what users are asking about your competitors
Output Format
# Google PAA Questions: "[KEYWORD]"
Queries run: [N] | Unique PAA questions: [N] | Featured snippet opportunities: [N] | Date: [DATE]
## PAA Question Bank (Sorted by Opportunity Score)
| # | Question | Intent | Appears For | Featured Snippet? | Score |
|---|---------|--------|------------|------------------|-------|
| 1 | "[question]" | INFORMATIONAL | [N] keywords | No | [0.XX] |
## High-Priority Questions (Use as H2/H3 in Content)
1. "[question]" — Answer in [X] words; [informational guide / comparison table / step-by-step]
2. "[question]"
## FAQ Section Builder
Questions suitable for FAQ schema markup:
1. Q: "[question]" — A: [1-sentence answer start]
## Content Angle Recommendations
- For "[primary keyword]" blog post: Use [N] of these PAA questions as headers
- For FAQ page: [N] questions qualify for FAQ schema
- For comparison page: [N] vs/comparison questions foundTroubleshooting
"includePeopleAlsoAsk" returns no results: Not all keywords trigger PAA boxes. Try adding "how", "why", "best", "what" as prefixes to force question-form SERPs. PAA questions are all branded (competitor names): High competitor brand presence on SERPs — separate branded vs. non-branded question sets in your content plan. Questions are too obscure: PAA expands dynamically based on prior searches — the questions returned reflect real user journeys; trust them even if obscure.

