Lexicon

Paul Truscott's Semantic SEO and GEO Lexicon defines 98 terms across semantic SEO, generative engine optimisation (GEO), AI search, entities and knowledge graphs, structured data, local SEO and financial technical analysis, including the terms Paul Truscott coined.

A

ADX (Average Directional Index): The Average Directional Index (ADX) is a technical indicator, developed by J. Welles Wilder in 1978, that measures the strength of a trend in a series of prices, without indicating the trend's direction.
AEO (Answer Engine Optimization): Answer Engine Optimization (AEO), industry terminology that emerged circa 2023 to 2024 with no single credited originator, is the set of strategies aimed at improving how visible and retrievable a brand or its content is to large language models, conversational agents, and AI-driven search engines, distinct from optimizing the AI systems themselves.
AI Mode: AI Mode is a Google Search feature, introduced in 2025, that presents an entirely AI-generated results page within a dedicated tab, functioning as a conversational AI assistant integrated directly into Google Search.
AI Overviews (AIO): AI Overviews is a Google Search feature that uses AI to generate a summary in response to a query, drawing on and citing multiple web sources, displayed prominently above traditional search results.
AI Résumé: An AI Résumé, coined by Jason Barnard and Kalicube, is the composite representation of a brand or individual across AI-generated answers and assistants, functioning as the AI-era equivalent of a Brand SERP: what AI systems collectively say about an entity when asked.
AI Search Optimization (AISO): AI Search Optimization (AISO), coined by Jason Barnard in 2023, is the process of tailoring a brand's entire digital ecosystem to be discoverable, trusted, and accurately represented by AI Assistive Engines like Google AI Mode, Bing Copilot, ChatGPT, and Perplexity.
AI SEO: AI SEO is one of several overlapping terms, alongside generative engine optimization (GEO) and answer engine optimization (AEO), for the practice of structuring content and managing online presence so it is discoverable and accurately represented by AI-powered search and answer systems, in addition to traditional search engines.
Assistive AI: Assistive AI, coined by Jason Barnard and Kalicube, refers to AI-powered engines and assistants, such as Google AI Mode, ChatGPT, and Perplexity, that act on a user's behalf by retrieving, synthesizing, and presenting information, including brand-related answers, rather than simply listing links for the user to evaluate.
Author Vectors / isAuthor: Author vectors, referenced in leaked Google internal documentation via the attribute isAuthor, are numerical representations that associate content with a specific author identity, enabling search systems to evaluate and aggregate the authority and expertise of individual content creators across their body of work.

B

BERT: BERT (Bidirectional Encoder Representations from Transformers) is a language model developed by Google that learns to represent text as a sequence of vectors by analyzing the context of words in both directions within a sentence, significantly advancing natural language processing.
Beta: Beta is a measure of a security's volatility relative to the overall market, indicating the degree to which the security's price tends to move in response to movements in a market index.
Bollinger Bands: Bollinger Bands are a type of statistical chart, developed by John Bollinger in the 1980s, that plot a moving average alongside upper and lower bands set at a specified number of standard deviations.
Brand SERP: A Brand SERP is the complete set of search results that appears when someone searches for a brand or individual by name, functioning as a public-facing representation of that entity's online identity and reputation.
Brand SERP Optimisation: Brand SERP optimization is the practice of influencing and improving the search results that appear when someone searches for a brand or individual by name, with the goal of presenting an accurate, favourable, and comprehensive representation of that entity.

C

Canonical: Canonical means conforming to the accepted standard, rule, or primary authoritative source within a given body of knowledge.
Canonical Source of Ground Truth: A canonical source of ground truth is the single most authoritative and verifiable reference point used to establish the accepted facts about a subject, against which other claims or data sources are validated.
Chunking: Chunking, in the context of information retrieval and AI, is the process of dividing a larger document or body of text into smaller, self-contained segments so that each can be individually indexed, retrieved, and processed by a language model.
CID (Customer ID): A CID is Google's unique numerical identifier for a business listing in Google Business Profile, used to reference a specific local business entity within Google's systems, including Maps and local search results.
Citation: A citation, in the context of AI-generated responses, is a reference to a specific external source that the AI system identifies and attributes as the origin of information it includes in its output.
Citation RSICoined by Paul Truscott: Citation RSI, coined by Paul Truscott in 2026, is a momentum framework, adapted from the Relative Strength Index used in technical analysis, that measures the speed and magnitude of recent change in an entity's citation or mention frequency across AI-generated answers.
Click-Independent Visibility: Click-independent visibility is the exposure and influence a brand achieves within search or AI-generated results without generating a click to the brand's own website.
Content Effort (contentEffort): Content effort is a quality signal, referenced in leaked Google internal documentation, that evaluates the degree of original work, research, and value a piece of content demonstrates beyond what is readily available elsewhere.
Contextual Vector: A contextual vector is a representation of a document or piece of content as a set of weighted terms and concepts that defines its position in a multi-dimensional meaning space relative to other content.
Conversational Search: Conversational search is a search interaction model in which users query a system using natural, dialogue-like language, and the system responds with contextually relevant answers, often maintaining context across multiple turns.
Correlation: Correlation is a statistical measure that describes the strength and direction of a relationship between two variables, expressed as a coefficient ranging from minus one, a perfect inverse relationship, to plus one, a perfect direct relationship.
Corroboration: Corroboration is the confirmation of a claim, fact, or account by evidence from one or more independent sources, strengthening confidence in its accuracy.
Cost-of-Retrieval: Cost-of-retrieval measures how much processing a search engine needs to extract a reliable answer from a document. Lower cost means the system can serve the answer faster and with higher confidence.

D

Digital PR: Digital PR, also known as online public relations, is the use of internet-based channels and publications to manage and enhance the public perception of a brand or individual, including securing earned editorial coverage that builds authority and visibility across search and AI systems.
Dow Theory: Dow Theory is a framework for understanding stock market behaviour derived from the writings of Charles Dow, based on the principle that market trends must be confirmed across independent indices and that trends persist until definitive signals indicate they have reversed.
Drawdown: A drawdown is the decline in value from a peak to a subsequent trough before a new peak is reached. It is used to measure the downside risk of an investment or strategy over a given period.

E

E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness): E-E-A-T is a framework, defined in Google's Search Quality Rater Guidelines and expanded from E-A-T in December 2022, used to evaluate the quality of web content.
Entity Disambiguation: Entity disambiguation is the process of determining which specific real-world entity a name or mention in text refers to, resolving ambiguity when the same name could refer to multiple distinct entities in a knowledge base.
Entity Reconciliation: Entity reconciliation is the process of matching and merging references to the same real-world entity across multiple independent data sources, confirming that separate records or mentions refer to a single, unified entity.
Entity Salience: Entity salience is a measure of how central or important a particular entity is to the overall meaning of a document, as distinct from merely being mentioned in it.
Entity Support and ResistanceCoined by Paul Truscott: Entity Support and Resistance, coined by Paul Truscott in 2026, is a framework, adapted from the support and resistance concept in technical analysis, that describes the floor and ceiling levels of trust an entity holds within AI systems.
Entity-Attribute-Value Model (EAV): The Entity-Attribute-Value model is a data structure in which each fact is stored as a triple of an entity, an attribute describing a property of that entity, and the value of that property, enabling flexible, extensible representation of entities with varying characteristics.

F

FAQPage Schema: FAQPage schema is a Schema.org markup type used to identify a page that contains a list of frequently asked questions and their answers, enabling search engines to display the questions and answers directly within search results.
Featured Snippet: A featured snippet is a summary of an answer to a user's query, extracted from a web page by the search engine and displayed at the top of the results page above the standard organic listings.

G

Generative SEO: Generative SEO refers to the practices aimed at improving how a brand's content is retrieved, summarized, and cited by generative AI systems when producing synthesized search answers.
GEO (Generative Engine Optimization): Generative Engine Optimization (GEO), coined at KDD 2024 by Aggarwal et al. of Princeton, Georgia Tech, the Allen Institute for AI, and IIT Delhi, is the practice of structuring content and managing online presence to improve how it is retrieved, summarized, and presented by generative AI systems when they answer user queries.
Grounding: Grounding, in artificial intelligence, is the process of connecting a model's generated output to verifiable external information sources, ensuring that its responses are anchored in factual, retrievable evidence rather than relying solely on learned patterns.

H

Hallucination: A hallucination, in artificial intelligence, is a response generated by a model that is presented as factual but is not grounded in the model's training data or any retrieved source, resulting in fabricated, inaccurate, or misleading output.
Holistic SEO: Holistic SEO is an approach to search optimization that treats a website's content, technical structure, and topical coverage as a single interconnected system, rather than optimizing individual pages or keywords in isolation.

I

Inference Optimization: Inference optimization is the process of improving the speed, efficiency, or cost of generating outputs from a trained AI model, without altering the model's underlying knowledge or architecture.
Information Responsiveness: A document is information-responsive when it answers the query directly, at the passage level, in a format the retrieval system can extract and present.

K

KGMID: A KGMID (Knowledge Graph Machine ID) is Google's internal unique identifier for an entity in its Knowledge Graph, typically formatted as /g/ followed by a string of characters, used to distinguish one entity from all others in the system.
Knowledge Graph: A knowledge graph is a knowledge base that uses a graph-structured data model to store and connect information about real-world entities, people, places, organizations, and concepts, and the relationships between them.
Knowledge Graph Engineering: Knowledge graph engineering is the practice of designing, building, and maintaining knowledge graphs, structured representations of entities and their relationships, through processes including data modelling, entity extraction, ontology design, and ongoing validation of the information stored within the graph.
Knowledge Panel: A Knowledge Panel is an information box displayed by Google alongside search results, presenting a structured overview of an entity, such as a person, organization, or place, using data drawn from Google's Knowledge Graph.
Knowledge Panel Management: Knowledge Panel Management is the practice of influencing and maintaining the accuracy, completeness, and presentation of an entity's Google Knowledge Panel by managing the underlying data sources that feed the Knowledge Graph.

L

LLM (Large Language Model): A large language model (LLM) is a type of artificial intelligence model trained on vast amounts of text data, designed to understand, generate, and process natural language by predicting the probability of sequences of words.
llms.txt: llms.txt is a proposed standard file, placed in a website's root directory, that provides structured information about the site and its content specifically for consumption by large language models and AI crawlers, similar in concept to how robots.txt communicates with search engine crawlers.
LocalBusiness Schema: LocalBusiness schema is a Schema.org markup type used to describe a physical business with a local presence, providing structured information such as address, opening hours, and contact details in a format that search engines can read and display.

M

MACD (Moving Average Convergence/Divergence): MACD is a trading indicator used in technical analysis that measures the relationship between two exponential moving averages of a security's price, designed to reveal changes in the strength, direction, momentum, and duration of a trend.
Macro-Micro Semantics: Macro semantics define the overall topic and source context of a page. Micro semantics operate at the sentence and phrase level: word sequences, co-occurrence patterns, predicate selection, and the distributional relationships between terms.
Mentions: A mention is any reference to a brand, individual, or entity within online content that does not include a hyperlink to the entity's own website, serving as a signal of recognition and relevance independent of traditional link-based authority.
MID (Machine ID): A MID (Machine ID) is a legacy entity identifier, formatted as /m/ followed by an alphanumeric string, originally assigned in Google's Freebase database and still used in parts of Google's Knowledge Graph to uniquely identify entities.
Moving Averages: A moving average is a calculation that smooths data by creating a constantly updated average of values over a specified number of periods, used to identify the direction and strength of a trend by filtering out short-term fluctuations.
MUM (Multitask Unified Model): MUM (Multitask Unified Model) is a Google AI model, announced in 2021, that is trained across 75 languages and multiple tasks simultaneously, designed to understand and generate language while also being capable of processing information across text and images.

N

Named Entity Recognition (NER): Named Entity Recognition (NER) is a subtask of information extraction that identifies and classifies named entities, such as people, organizations, locations, and products, within unstructured text.
Neural Matching: Neural matching is a technique used by search engines to understand the relationship between words and concepts in queries and documents by using neural networks to match the intent of a query with relevant content, even when exact keywords are not present.

O

On-SERP SEO: On-SERP SEO is the practice of optimizing a brand's presence within the search results page itself, through knowledge panels, featured snippets, and other SERP features, rather than solely pursuing organic link rankings.
Online Reputation Management (ORM) for AI: Online reputation management is the practice of monitoring, influencing, and maintaining the perception of an individual or brand across internet platforms, including managing the accuracy and favourability of information that appears in search results and AI-generated responses.
Organization Schema: Organization schema is a Schema.org markup type used to describe an organization, such as a company, charity, or institution, and its properties, including name, logo, contact information, and relationships to other entities.

P

Passage Indexing: Passage indexing is a search engine capability, announced by Google in 2020, that allows individual passages within a longer web page to be independently indexed and ranked in response to a query, rather than evaluating only the page as a whole.
Paul Truscott Coined TermCoined by Paul Truscott: A Paul Truscott Coined Term, coined by Paul Truscott in 2026, is a specific word or phrase originated by Paul Truscott to precisely name and define a new concept or framework at the intersection of semantic SEO, generative engine optimization, and technical analysis.
People Also Ask (PAA): People Also Ask is a Google Search feature that displays a list of related questions beneath the main results, each expanding to reveal a brief extracted answer and a link to its source page.
Position Sizing: Position sizing is the process of determining how much capital to allocate to a single trade or investment, based on the level of risk acceptable relative to the total portfolio, to manage exposure and protect against disproportionate loss.
Prominence: Prominence is a measure of how well-known and widely recognised an entity is, based on the breadth and depth of information available about it across the web, including references in directories, editorial coverage, reviews, and other independent sources.

Q

Query Fan-Out: Query fan-out is the process by which an AI search system breaks a single user query into multiple related sub-queries, retrieves results for each independently, and synthesizes them into one combined answer.

R

RankBrain: RankBrain is a machine learning-based component of Google's search algorithm, introduced in 2015, that helps process and interpret search queries, particularly unfamiliar or ambiguous ones, by converting them into mathematical vectors that the system can use to find relevant results.
RankEmbed: RankEmbed is a Google system, referenced in leaked internal documentation, that uses embedding-based representations to evaluate and rank content by comparing the vector similarity between queries and documents.
Rate of Change (ROC): Rate of change is a momentum indicator that measures the percentage change in price between the current price and the price a specified number of periods ago, expressing the speed at which a value is changing over time.
Regression to the Mean: Regression to the mean is the statistical phenomenon whereby extreme measurements or outcomes tend to be followed by values closer to the average, not because of any corrective force, but because extreme observations are partly the result of random variation that is unlikely to repeat.
Retrieval-Augmented Generation (RAG): Retrieval-Augmented Generation (RAG) is a technique that enables large language models to retrieve and incorporate information from external data sources at the time of generating a response, rather than relying solely on their pre-existing training data.
Risk/Reward Ratio: The risk/reward ratio is a measure that compares the potential loss of a trade or investment to its potential gain, used to assess whether the expected return justifies the risk being taken.
RSI (Relative Strength Index): The Relative Strength Index (RSI) is a momentum oscillator used in technical analysis that measures the speed and magnitude of recent price movements on a scale of zero to one hundred.

S

Schema Markup: Schema markup is a standardized vocabulary of tags, maintained by Schema.org, that can be added to a web page's code to help search engines understand the meaning and structure of the content, enabling richer presentation in search results.
Semantic SEO: Semantic SEO is the practice of structuring content around meaning, entities, and context, rather than isolated keywords, so that search engines can understand relationships between concepts and match content to the underlying intent of a query.
SGE (Search Generative Experience): Search Generative Experience (SGE) was Google's original name, coined by Google and introduced in May 2023, for its AI-powered search feature that generated synthesized conversational answers above traditional search results, before being rebranded as AI Overviews and AI Mode.
Share of Voice: Share of voice is a measurement of the proportion of total visibility or media presence a brand holds within a defined market or category, compared to its competitors, over a given time period.
Sharpe Ratio: The Sharpe ratio is a measure of risk-adjusted return, calculated as the difference between the return of an investment and the risk-free rate divided by the standard deviation of returns, used to evaluate how much excess return is received for each unit of risk taken.
Site Authority (siteAuthority): Site authority is a quality signal, referenced in leaked Google internal documentation, that represents an overall assessment of a website's credibility, trustworthiness, and standing as a source within its subject area.
Source Context: Source context refers to the established topical identity and authority of a website or content source as understood by a search engine, determining how that source's content is interpreted and weighted for relevance within a given subject area.
Standard Deviation: Standard deviation is a statistical measure of the amount of variation or dispersion in a set of values, indicating how far individual data points typically deviate from the mean of the dataset.
Stochastics: A stochastic oscillator is a momentum indicator used in technical analysis that compares a security's closing price to its price range over a specified period, indicating where the current price sits relative to its recent high and low.
Structured Data: Structured data is information that is organized in a predefined, machine-readable format, such as JSON-LD, Microdata, or RDFa, allowing software systems to reliably identify and interpret specific facts, entities, and relationships within a document.
Sub-query Decomposition: Sub-query decomposition is the technique of breaking a complex search query into smaller, more specific component questions so that each can be answered or retrieved separately before being recombined into a unified response.
Support and Resistance: Support and resistance are price levels on a chart where the forces of supply and demand tend to meet. Support represents a level where buying interest is strong enough to prevent further decline.
SXO (Search Experience Optimization): Search Experience Optimization (SXO), a framing associated with Shaun Anderson of Hobo Web as a proposed successor term to SEO, centres on optimizing the entire user search experience, relevance, usability, and satisfaction, rather than ranking signals alone, reflecting search engines' growing use of user-experience data in ranking.

T

Topical Authority: Topical authority is the degree to which a website or content creator is recognised as a credible and comprehensive source of information within a specific subject area, built through consistent depth and breadth of coverage across that topic.
Topical Map: A topical map is a structured plan of all the subjects, subtopics, and their relationships that a website needs to cover in order to demonstrate comprehensive authority within a defined topic area.
Training Data / Training Corpus: Training data is the collection of examples or information used to teach a machine learning model to recognize patterns and make predictions, forming the foundation of the model's knowledge and capabilities.

V

Vector Embeddings / Word Embeddings: A word embedding is a representation of a word as a numerical vector in a continuous vector space, where words with similar meanings are positioned closer together, enabling machines to process language based on semantic relationships rather than exact string matches.
Visibility Bollinger BandsCoined by Paul Truscott: Visibility Bollinger Bands, coined by Paul Truscott in 2026, is a framework, adapted from the Bollinger Bands indicator in technical analysis, that plots a moving average of an entity's tracked citation or visibility frequency alongside upper and lower bands set at a specified number of standard deviations.
Visibility DrawdownCoined by Paul Truscott: Visibility Drawdown, coined by Paul Truscott in 2026, is a framework, adapted from the drawdown concept in technical analysis, that measures the decline in an entity's citation or visibility level from a prior peak to a subsequent trough.

Z

Z-Score: A Z-score is a statistical measurement that expresses how many standard deviations a given data point is from the mean of a dataset, allowing values from different scales or distributions to be compared on a standardised basis.
Zero-Click Search: A zero-click search is a query that is fully resolved on the search engine results page itself, with the user obtaining their answer without clicking through to any external website.