How do you identify which content should be your 'cornerstone' for AI visibility?

Here’s my framework for identifying cornerstone opportunities:

Cornerstone Criteria Matrix:

Process:

1. List your top 15-20 topic candidates
2. Score each on the matrix
3. Top 5-7 become cornerstone candidates
4. Evaluate existing content for each
5. Decide: upgrade existing or create new

Key insight:

The best cornerstones sit at the intersection of:

• What your business needs to be known for
• What your audience is searching/asking
• Where you have genuine expertise
• Where AI answers are currently weak

The “upgrade vs create new” decision framework:

Upgrade existing when:

Create new when:

The 80/20 rule:

In my experience, 80% of cornerstone opportunities are upgrades, not new creation. Your existing content that ranks or gets traffic is your starting point.

Our process:

1. Export top 100 pages by traffic
2. Categorize by topic cluster
3. Identify best performer per cluster
4. Evaluate as cornerstone candidate
5. Upgrade rather than replace

Structuring content around cornerstones for AI:

The hub-and-spoke model:

CORNERSTONE (comprehensive, 3000+ words)
├── Supporting article 1 (specific subtopic)
│   └── Links to cornerstone + related articles
├── Supporting article 2 (specific subtopic)
│   └── Links to cornerstone + related articles
├── Supporting article 3 (case study)
│   └── Links to cornerstone
└── Supporting article 4 (FAQ)
    └── Links to cornerstone

Why this helps AI visibility:

1. Clear hierarchy - AI understands what’s important
2. Internal links - Signals topic authority
3. Comprehensive coverage - Multiple angles covered
4. Entry points - Supporting content captures long-tail

Practical implementation:

For each cornerstone:

• Identify 8-15 supporting topics
• Ensure each supports article exists
• Add clear internal linking both ways
• Update cornerstone to reference all supporting content

The cornerstone becomes your “everything about X” page.

Data-driven cornerstone identification:

Metrics that identify cornerstone candidates:

Analysis query:

Export pages with:

• 500+ monthly sessions
• Ranks position 1-20 for primary keyword
• 10+ backlinks
• 2+ minute avg time on page

These are cornerstone candidates.

Additional filters:

• Topic relevance to business
• Content quality assessment
• Update/refresh potential

Don’t just go by traffic. A high-traffic page on a tangential topic isn’t a good cornerstone.

Audit framework for 400+ posts:

Phase 1: Categorization (2-3 days)

1. Export all posts (title, URL, topic)
2. Assign primary topic to each
3. Group into 10-15 topic clusters
4. Count posts per cluster

Phase 2: Performance analysis (2-3 days)

For each cluster:

• Which posts get traffic?
• Which posts rank?
• Which posts have backlinks?
• Which posts get AI citations?

Phase 3: Cornerstone selection (1-2 days)

For each of your top 5-7 clusters:

• Identify best-performing post
• Evaluate cornerstone potential
• Document gaps to address

Phase 4: Action planning (1 day)

Output:

Prioritized list of 5-7 cornerstone opportunities with clear action plans.

AI-specific cornerstone optimization:

What makes cornerstone content AI-citation worthy:

1. Comprehensive coverage

   • Addresses all aspects of topic
   • Anticipates follow-up questions
   • Provides depth and breadth

2. Clear structure

   • Obvious hierarchy (H1, H2, H3)
   • TL;DR at top
   • FAQ section at bottom
   • Tables for comparisons

3. Information gain

   • Original data or research
   • Expert perspectives
   • First-hand experience
   • Unique frameworks

4. Authority signals

   • Expert author byline
   • Citations to reputable sources
   • Regular updates with timestamps

Cornerstone checklist:

• Answers the core question in first 100 words
• Covers 10+ subtopics comprehensively
• Includes original data/insights
• Has clear structure with headers
• Links to supporting content
• Named expert author
• Updated within 6 months

This discussion has given me exactly what I needed. Here’s my plan:

Phase 1: Audit (Week 1)

• Categorize 400+ posts into topic clusters
• Identify top performers per cluster
• Analyze AI citation status

Phase 2: Selection (Week 2)

• Apply criteria matrix to top candidates
• Select 5-7 cornerstone opportunities
• Determine upgrade vs create for each

Phase 3: Prioritization (Week 2) Score each by:

• Business alignment (25%)
• Search demand (20%)
• Expertise advantage (20%)
• Existing assets (15%)
• AI opportunity (20%)

Cornerstone optimization plan (Weeks 3-8):

For each cornerstone:

1. Audit current content quality
2. Identify gaps vs comprehensive coverage
3. Add information gain elements
4. Implement AI-friendly structure
5. Create/update supporting content
6. Build internal linking structure

Success metrics:

• Rankings for topic keywords
• Organic traffic to cluster
• AI citation rate for topic
• Backlinks to cornerstone

Key insight:

Start with what’s already working (upgrades), focus on topics where AI answers are weak, and build comprehensive clusters around each cornerstone.

Thanks everyone for the practical frameworks.