Creation
A person or system produces a fact, claim, image, record, policy or description. Provenance and ownership begin here.
From evidence to action
The Search Discovery Stack
A twelve-layer operating model for following information from creation through customer or agent action.
Create, publish and discover
Publication
The material becomes available through a page, profile, feed, API, file or database. Visibility depends on access conditions.
Discovery
Crawlers, links, submissions, sitemaps and connected datasets expose the material to systems that may process it.
Index, recognize and retrieve
Indexing
A service processes material into a search index, vector store, knowledge base or other searchable representation.
Recognition
Names, identifiers, relationships and context are used to determine which real-world entity the material describes.
Retrieval
The system selects documents, passages, fields or records relevant to the request. Poor retrieval can exclude the best source before generation starts.
Generate, attribute and present
Generation
A model turns instructions and context into language, comparisons, recommendations or actions.
Attribution
The product may attach links or citations, but the citation may support only part of the generated statement.
Presentation
Interface rules determine order, prominence, number of choices, visual treatment and whether the user continues to a source.
Evaluate, select and act
Reliability
Freshness, source quality, retrieval quality and generation fidelity determine whether the answer is useful and accurate.
Selection
The user or agent chooses among the visible options. Omitted alternatives may never be considered.
Action
The journey produces a call, visit, booking, purchase, subscription, tool execution or no action at all.
The stack is larger than any one acronym
SEO is strongest in publication, crawling, indexing, relevance and presentation. REG concentrates on recognition and retrievable facts. AEO structures direct answers. GEO considers usefulness in generated responses. RAG is an architecture inside retrieval and generation. Reliability frameworks such as NIST span governance, measurement and risk across the entire system.
The stack makes those relationships visible without pretending that the boundaries are perfectly rigid.
Questions for every layer
- Creation: Who produced the information and what evidence supports it?
- Publication: Is it public, current, canonical and understandable?
- Discovery: Can legitimate crawlers reach it through normal links and responses?
- Indexing: Is the correct version eligible and present in relevant indexes?
- Recognition: Can the system distinguish the entity from similar names and outdated records?
- Retrieval: Does the passage contain the needed fact and its limitation?
- Generation: Does the output remain faithful to the retrieved evidence?
- Attribution: Does the cited source actually support the claim?
- Presentation: How many choices and links are visible?
- Reliability: What could be stale, incomplete or fabricated?
- Selection: Why would the user consider this option?
- Action: Can the user or agent complete the next step safely?
Primary material behind this analysis
AI Features and Your Website
Official explanation of AI Overviews, AI Mode, query fan-out, eligibility and controls.
Publishers and Developers FAQ
Official information about OAI-SearchBot, GPTBot, referral tracking and agent accessibility.
NIST AI Risk Management Framework: Generative AI Profile
Cross-sector guidance for governing, mapping, measuring and managing generative-AI risk.
Open the original source at National Institute of Standards and Technology →
Continue through the discovery stack
The Hidden Radius Model
The original Hidden Radius framework for mapping evidence, discovery, recognition, retrieval, generation, reliability and action beneath modern search.
Search and AI Visibility Outcomes
A measurement framework separating discoverability, indexability, recognition, retrieval, answerability, attribution, representation, selection and business action.
How SEO, GEO, AEO, AIO, REG, RAG and LLMs Work Together
A complete explanation of where the major modern-search terms overlap across publication, identity, retrieval, generation, presentation and outcomes.
Modern Search and AI Discovery Business Examples
Detailed examples showing how the discovery stack changes for local services, ecommerce, publishers, professional firms, documentation, RAG and agents.