Video, Wikipedia, Reddit - which content types actually get cited by AI platforms?

Here’s the comprehensive citation data:

Overall content type citation rates:

Platform-specific preferences:

Why this matters: A single content strategy won’t work across all AI platforms. You need platform-specific approaches.

Key insight: “Specialized sites” (48-77%) represents industry-specific sources. For your niche, there are likely dominant sources you should appear on.

Wikipedia is complicated. Here’s the reality:

Why Wikipedia matters for ChatGPT:

• 22% of LLM training data comes from Wikipedia
• ChatGPT cites Wikipedia in 47.9% of its top sources
• It’s treated as neutral, authoritative baseline

Can you get a Wikipedia page? Notability requirements:

• Significant coverage in reliable sources
• Independent of the subject
• Not press releases or paid content

If you can’t get a page:

1. Wikidata - Lower bar, still valuable
2. Citations on existing pages - Get mentioned on industry pages
3. Wikipedia-style content - Structure your content like Wikipedia

What works:

• Neutral, factual tone
• Extensive citations
• Comprehensive coverage
• Regular updates
• Clear structure

The honest truth: Most businesses won’t get Wikipedia pages. Focus on:

1. Getting mentioned on existing Wikipedia pages
2. Creating Wikipedia-style content on your own site
3. Building presence where you CAN appear

Reddit’s influence is platform-dependent but growing:

Reddit citation by platform:

Why Perplexity loves Reddit:

• Peer-to-peer authenticity
• Real-world experiences
• Diverse perspectives
• Updated discussions

Reddit strategy (if appropriate for your brand):

1. Genuine participation in relevant subreddits
2. Answer questions helpfully (no promotion)
3. Share expertise, not links
4. Build karma and account age

Warning signs - avoid:

• Obvious promotion
• New accounts pushing products
• Link dropping
• Astroturfing (will get you banned and hurt brand)

Better approach: Encourage genuine customers to share experiences. Monitor brand mentions. Engage authentically when appropriate.

Industry relevance: Gaming, tech, finance - Reddit very influential B2B enterprise, healthcare - Less relevant

Citation patterns by industry:

Health/Medical:

• NIH: 39%
• Healthline: 15%
• Mayo Clinic: 14.8%
• Cleveland Clinic: 13.8%
• YouTube: 28%

Finance:

• YouTube: 23%
• Wikipedia: 7.3%
• LinkedIn: 6.8%
• Investopedia: 5.7%

E-commerce:

• YouTube: 32.4%
• Shopify: 17.7%
• Amazon: 13.3%
• Reddit: 11.3%

Gaming:

• YouTube: 93%
• Reddit: 78%
• Fandom wikis: 26.7%
• Steam: 11%

Key insight: Your industry has dominant sources. Identify them. Get presence on them.

Action: Ask AI questions about your industry. Note which sources get cited. Those are your targets.

Don’t abandon blog content yet. Here’s nuance:

Blog content CAN get cited when it:

• Contains original research/data
• Uses expert-led, bylined content
• Has strong E-E-A-T signals
• Is structured for AI extraction
• Covers topics YouTube/Reddit don’t

Blog content struggles when it:

• Restates common knowledge
• Lacks author expertise signals
• Uses dense, unstructured format
• Competes with video for how-to queries
• Has no unique data or perspective

The “specialized sites” category: That 48-77% of citations going to specialized sites? Many are business blogs with:

• Industry expertise
• Original data
• Proper structure
• Authority signals

Your strategy: Don’t choose blog OR video. Choose blog AND video for different purposes.

Blog: Data, analysis, comparisons, technical deep-dives Video: How-to, demonstrations, explanations, reviews

Here’s how to think about platform-specific strategy:

ChatGPT optimization:

• Focus: Authority, brand recognition, Bing visibility
• Content: Wikipedia-style structure, comprehensive
• Priority: Build brand mentions across authoritative sources

Perplexity optimization:

• Focus: Fresh content, clear answers, community presence
• Content: Real-time updates, Reddit engagement
• Priority: Be where Perplexity looks (news, community, fresh content)

Google AI Overview:

• Focus: Traditional SEO + AI structure
• Content: Well-ranked, structured pages
• Priority: Maintain Google rankings, add AI-friendly structure

Multi-platform approach: Create core content, then:

• Structured blog post (Google AI)
• Video version (YouTube/all platforms)
• Community discussion participation (Perplexity)
• Press coverage for authority (ChatGPT)

One topic, multiple formats, multiple platforms.

Here’s how I pivoted our content strategy:

Before (blog-heavy):

• 80% blog posts
• 10% video
• 10% other

After (diversified):

• 50% blog (restructured for AI)
• 30% video (YouTube focus)
• 15% data/research
• 5% community participation

Changes made:

1. Every blog post now has video version (or vice versa)
2. Added original research to key topics
3. Created YouTube strategy for how-to content
4. Genuine Reddit/community participation program
5. Press coverage for authority building

Results after 6 months:

• AI citations: Up 340%
• Video citations: 45% of total AI visibility
• Blog citations: 35% of total (up from near zero)
• Traffic from AI: 12% of organic

The lesson: It’s not about abandoning blog content. It’s about:

1. Restructuring blog for AI extraction
2. Adding video for demonstration content
3. Building multi-platform presence

Great insights. Here’s my revised strategy:

Content type allocation (going forward):

Platform-specific approach:

For ChatGPT:

• Build brand authority through PR
• Create Wikipedia-style comprehensive content
• Focus on Bing visibility

For Perplexity:

• Fresh, regularly updated content
• Genuine community participation
• Clear, extractable answers

For Google AI:

• Maintain strong SEO
• Add AI-friendly structure
• Implement schema markup

Format decisions:

• How-to content → Video first, blog version second
• Data/research → Blog first, video summary second
• Comparisons → Blog with tables, video for detailed review
• Definitions → Blog with Wikipedia-style structure

Tracking:

• Use Am I Cited to monitor by content type
• Track which formats get cited where
• Iterate based on data

Not abandoning blog, but diversifying and restructuring.

Thanks for the data and perspectives!