NRLC Doctrine

Pages are organizational containers for chunks. They do not exist as retrievable units in AI systems.

AI systems extract chunks from pages. Pages function as delivery mechanisms, not retrieval targets.

Page structure exists to organize chunks and control chunk boundaries, not to rank as documents.

Page-level metrics like rankings and traffic are outputs of container performance, not chunk performance.

Optimizing pages as containers requires understanding how chunk boundaries affect retrieval.

This doctrine contradicts traditional SEO that treats pages as ranked documents. Prechunking operates at the chunk level.

Chunks are the actual assets that AI systems retrieve and cite. They are the units of value in AI search.

Chunk quality determines retrieval success. Poor chunks are ignored. Good chunks are cited.

Chunks must be engineered as standalone assets. They cannot depend on page context for accuracy.

Chunk inventory is the asset portfolio. Missing chunks cause incomplete answers and competitor citation.

Chunk performance is measured through citation rates and answer inclusion, not page rankings.

This doctrine requires treating chunks as first-class assets, not as page components.

Retrieval algorithms determine which chunks are available for AI answers. Ranking algorithms determine which pages appear in search results.

Retrieval happens before ranking in AI systems. Chunks must pass retrieval tests before ranking signals matter.

A page can rank first but have zero chunks retrieved if its chunks fail retrieval tests.

A page can rank tenth but have multiple chunks retrieved if its chunks pass retrieval tests.

Prechunking optimizes for retrieval, not ranking. Ranking optimization is separate and secondary.

This doctrine explains why high-ranking pages are often ignored by AI systems. They fail at retrieval, not ranking.

Facts must remain accurate when extracted from context. Truth cannot depend on surrounding text for correctness.

Isolation testing validates that facts are self-contained. Facts that mutate when isolated are not true croutons.

Truth engineering requires explicit statements, not implied meanings. Implied truth does not survive extraction.

Facts must be stated clearly enough to be understood without explanation. Context-dependent truth is not retrievable truth.

This doctrine requires discipline in content creation. Narrative truth must be transformed into declarative truth.

Truth that survives isolation is truth that can be trusted when cited by AI systems.

These doctrines are non-negotiable. Practices that contradict them are not prechunking SEO.

Doctrine enforcement requires:

• Auditing content for compliance with container and asset separation
• Validating that chunks can survive isolation
• Measuring retrieval performance, not just ranking performance
• Rejecting practices that optimize pages at the expense of chunks
• Prioritizing chunk quality over page metrics

Doctrine violations cause prechunking failures. Compliance ensures retrieval success.

All prechunking work must align with these doctrines. No exceptions.

• Prechunking SEO Overview
• Core Concepts
• Prechunking Workflow
• Failure Modes