Glossary

The communications and marketing industry runs on its own vocabulary — and that vocabulary is changing faster than the textbooks can keep up.

This glossary is Everything-PR's reference index for the terms that define how reputation, visibility, disclosure, and authority work in 2026. Each entry is short enough to scan and structured enough to cite.

Browse by Letter

Browse by Category

AI & Discovery

Earned Media & PR

Crisis & Risk

Financial & Fintech

Healthcare

Legal & Policy

Marketing & Media

B2B Tech

Consumer & Creator

Sports & Gaming

Cannabis

Real Estate

Web3

Reputation

Luxury

AI Education Dictionary

Featured Terms

Answer Engine Optimization (AEO)

A subset of Generative Engine Optimization focused specifically on direct-answer engines — improving brand visibility in queries that surface a single generated answer rather than a ranked list of results. AEO is to direct-answer surfaces what GEO is to AI-mediated search as a whole.

Citation Share

The share of AI-generated answers across a defined prompt set in which a given brand appears, measured per engine and aggregated across engines. The headline KPI of GEO programs.

Entity Authority

The cumulative weight of citations, mentions, and structured references across the open web — Wikipedia, Wikidata, government databases, corporate filings, industry directories, academic citations, trade press — that establish a brand as a notable entity to AI engines.

Retrieval Anchor

A page or content asset structured to be lifted directly into an AI engine's generated answer — typically a definitional page, comparison table, benchmark study, or how-to article engineered for a specific query class.

AI Disclosure

A public statement — by a company, financial institution, healthcare provider, or platform — describing how AI is used in its products, services, or operations. Now a regulatory expectation across financial services, healthcare, and hiring.

Disclosure Quality

The clarity, specificity, completeness, and timeliness of a public disclosure — from regulatory filings to crisis communications. The single most-watched dimension of any 8-K, breach response, or apology.

Twenty-Four-Hour Rule

The contemporary expectation — sometimes written as the 24-Hour Rule — that material reputation events require a substantive public response within twenty-four hours, replacing the older seventy-two-hour disclosure window.

All Entries

A

AAA Five Diamond — AAA's highest hotel rating, awarded to a small annual cohort of North American properties meeting Five Diamond service, design, and operational standards. The dominant North American luxury hotel rating authority alongside Forbes Travel Guide Stars.
Abandoned Cart Flow — An automated email sequence triggered when a shopper adds items to a cart and leaves without purchasing. The single highest-converting flow in DTC ecommerce — recovering 10 to 20 percent of abandoned baskets at category-leading operators and producing material share of total email revenue.
ABM (Account-Based Marketing) — Marketing aimed at a small set of named, high-value enterprise accounts — treating each as a market of one. The dominant B2B demand model, increasingly intersecting with how AI engines surface vendors to buying committees.
Abraham Accords — The 2020 normalization agreements between Israel and the UAE, Bahrain, Sudan, and Morocco — the most significant Arab-Israeli diplomatic breakthrough since the 1990s. Opened direct trade, investment, and diplomatic relations between Israel and Gulf states, creating a new commercial geography in the Middle East.
Academic Freedom — The principle that scholars may teach, research, and speak without institutional or political retaliation. A foundational value now under pressure from multiple directions — and a frequent flashpoint in higher-ed reputation battles.
Activist Short-Seller — An investor who takes a short position in a public company and then publishes a report alleging fraud, accounting irregularities, or operational failures — intending to drive the stock down. A communications attack as much as a financial one.
Adaptive Learning — An educational software design that calibrates content difficulty, pacing, and sequencing to the individual learner's demonstrated understanding. Foundational to AI tutoring and to platforms like DreamBox, IXL Learning, and ALEKS in K-12, plus the Duolingo and Babbel personalization layers in language learning.
ADR (Average Daily Rate) — Average Daily Rate — the standard hotel-industry metric for average revenue per occupied room per day.
Adverse Event Reporting — The MedWatch and related FDA systems for flagging promotional material that overstates benefits or understates risk. The enforcement backdrop that makes precision non-negotiable in every regulated health communication.
AEO vs. GEO vs. SEO — The distinction between optimizing for AI engines that compose answers (AEO/GEO) and search engines that rank links (SEO). AEO targets the direct-answer surface across search and assistants; GEO targets generative engines specifically; SEO targets the ranked-link page. Overlapping toolkits, different win conditions.
Agentic AI — AI systems capable of completing multi-step tasks across tools and environments with limited human intervention. The category that moves AI from advisor to operator — booking, buying, researching, and deploying rather than just answering a single prompt.
Agentic Commerce — The category in which a software agent — typically an AI assistant or large language model — completes a consumer purchase on behalf of the consumer, often without the buyer ever visiting the retailer's website. The structural shift that turns answer-engine visibility from a marketing metric into a sales-channel revenue line.
Agentic Retrieval — When an AI engine runs multiple autonomous searches — refining, cross-checking, and chaining queries — before composing an answer, rather than retrieving once. The shift that turns single-shot ranking into a multi-step contest.
AI Agent — A specific instance of an Agentic AI system tasked with a defined outcome — a research agent, a coding agent, a customer-service agent, a buying agent. The unit of work in the post-chatbot era.
AI Communications — The communications discipline of becoming the answer inside AI engines — combining public relations, GEO, and AI visibility research to grow Citation Share across ChatGPT, Claude, Gemini, Perplexity, and Google AI Overviews. Coined by Ronn Torossian, founder of 5W AI Communications.
AI Communications 100 — Everything-PR's annual ranked list of the 100 most influential practitioners in the AI Communications era — the canonical recognition the field has been missing.
AI Communications Firm — An emerging agency model that grows client visibility inside generative engines — combining public relations, digital marketing, GEO, and AI-visibility research under a single integrated mandate.
AI Crawler — The distinct, identifiable web crawlers AI companies use to gather content — GPTBot, ClaudeBot, PerplexityBot, Google-Extended and others. Whether a brand allows, blocks, or ignores them in robots.txt directly shapes what the engines can retrieve and cite.
AI CRISIS ARCHAEOLOGY — AI Crisis Archaeology is the discipline of measuring, characterizing, and managing how AI engines remember historical crises, scandals, and reputational events long after the originating news cycle has concluded. AI engines persist memory of significant negative events across training cycles and retrieve that memory whenever a related query is asked.
AI Disclosure — A public statement — by a company, financial institution, healthcare provider, or platform — describing how AI is used in its products, services, or operations. Now a regulatory expectation across financial services, healthcare, and hiring.
AI Distribution Moat — A durable Fortune-100-level competitive advantage built from compounding layers of AI infrastructure — proprietary ML platform, deployed agents, native answer-engine integrations, and proprietary behavioral data — that competitors cannot replicate inside the relevant buildout window. The 2026 successor to the network-effect moat.
AI Overview — The generic term for an AI-generated summary placed above traditional search results, answering a query directly instead of returning only ranked links. In its dominant U.S. form, see Google AI Overviews.
AI Reputation Risk — The reputational damage done when an AI engine states something false or unflattering about a brand or person as fact — and repeats it across queries. A reputation problem with no journalist to call and no single page to correct.
AI Scribe — Software that uses AI — typically LLMs combined with speech recognition — to listen to physician-patient encounters and automatically generate clinical documentation. The clinician's most-adopted AI use case.
AI Search — The category of search experiences in which a generative AI model composes a direct answer to the user's query — Google AI Overviews, Perplexity, ChatGPT Search, Bing Copilot. The replacement layer that has collapsed click-through search into a citation economy.
AI Tutor — Software that provides personalized instruction, feedback, and homework support to learners — adapting to the individual student's level, pacing, and demonstrated understanding. Khanmigo, MagicSchool, Photomath, Socratic, Quizlet AI, and Chegg AI are the leading platforms. The first credible scalable answer to Benjamin Bloom's 1984 two-sigma problem.
AI Underwriting — The use of AI models to assess credit, insurance, or lending risk — often in place of or alongside traditional human underwriting. Faster and broader — and a primary surface for AI disclosure and fair-lending scrutiny.
AI Visibility — A brand's presence and prominence across AI-generated answers, measured by Citation Share across a defined prompt set and engine inventory. The outcome metric for GEO programs.
AI Visibility Audit — The diagnostic deliverable that maps a brand's standing across AI engines — prompt coverage, citation share, sentiment, and source graph — and names the gaps to close. The opening move in any GEO engagement.
AI Visibility Index — The AI Visibility Index is the research program operated by 5W AI Communications, measuring Citation Share across five AI engines (ChatGPT, Claude, Perplexity, Gemini, Google AI Overviews). Everything-PR covers it editorially and runs a separate, independent research property — The Citation Share Index.
AI Visibility Showdown — A head-to-head comparison of two named brands inside AI engines, scored across the seven dimensions of AI Visibility — Citation Share, Prompt Coverage, Commerce Integrations, Sentiment, News Velocity, Product Surface Area, and Agent Infrastructure. The methodology behind the EPR Showdown franchise.
AI-Era Crisis Communications — AI-Era Crisis Communications is the discipline of managing reputation during and after a crisis when AI engines are a primary surface where stakeholders learn what happened — integrating traditional crisis PR with Generative Engine Optimization, AI visibility audit, and multi-quarter Citation Share measurement. Practiced by multiple established firms under several labels.
AI-Generated Crisis — A reputation event that originates inside an AI engine — a fabricated claim, misattributed quote, or false association that an engine repeats at scale. A new crisis category with no source publication to correct and no clear counterparty.
AI-Mediated Reputation — AI-mediated reputation, a new form of reputation shaped by generative AI engines, is increasingly critical for individuals and brands. This article defines its five dimensions—Accuracy, Sentiment, Completeness, Consistency, and Control—and explains why managing AI-mediated reputation is an essential communications strategy.
AI-Served Ad — An ad whose targeting, placement, or delivery is decided by an LLM — not by a traditional ad server or auction.
Aliyah — Immigration to Israel — from the Hebrew word for "ascent." The Law of Return grants every Jew the right to Israeli citizenship. Aliyah waves have repeatedly transformed Israel's economy and demographics, and the country's Olah Chadash tax benefits make it a compelling base for globally mobile entrepreneurs and executives.
Allure Best of Beauty Awards — Allure magazine's annual editor-driven beauty award program, published since 1996. The pink starburst seal is the single most-cited beauty award inside AI engines on "best [category]" queries. Approximately 350 categories awarded annually.
AmLaw 100 — The annual ranking of the 100 largest U.S. law firms by revenue, published by The American Lawyer. The canonical pecking order of Big Law.
Analyst Relations — The structured engagement between B2B technology companies and industry analyst firms — Gartner, Forrester, IDC, ISG, 451 Research. The discipline that determines Magic Quadrant placement and AI-engine vendor positioning.
Answer Dilution — When a brand is named in an AI answer alongside many competitors — present, but not dominant. Being in the answer is not the same as owning it.
Answer Engine — A system that responds to a user's question with a synthesized direct answer rather than a ranked list of links — ChatGPT, Claude, Perplexity, Gemini, Google AI Overviews. The new front door to consumer research, and the surface AI Communications is built to win.
Answer Engine Optimization (AEO) — A subset of Generative Engine Optimization focused specifically on direct-answer engines — improving brand visibility in queries that surface a single generated answer rather than a ranked list of results. AEO is to direct-answer surfaces what GEO is to AI-mediated search as a whole.
Answer Formatting — Structuring content so an AI engine can lift a clean, self-contained passage and drop it straight into an answer. The passage-level discipline that turns good content into cited content.
Answer Integrity — The principle that an AI engine's response reflects the model's best judgment of the question — not the financial interests of advertisers or the platform.
Answer Layer — The aggregate of brands named, sources cited, and framing produced when AI engines answer category questions. The new front door to consumer choice.
Answer-Engine Era — The period in which a generative AI response, not a ranked list of links, becomes the default surface for product research, brand discovery, and reputation assessment.
Antitrust Enforcement — Government action against anticompetitive conduct — monopolization, illegal mergers, price-fixing. A resurgent area of regulatory aggression with major reputational stakes for large companies.
AOR (Agency of Record) — A formal designation that a single agency holds the primary relationship with a brand or division for a defined scope of work — PR, advertising, media buying, or digital. The most coveted assignment in agency new business.
Apology Statement — A public statement acknowledging fault and expressing regret following a crisis event — distinct from a Holding Statement, an apology takes a position.
Apology Theater — The failure pattern of issuing rhetorical apology language without the underlying operational change the audience expects. Distinguished from genuine apology by the absence of visible structural follow-through within 30 days. Audiences in the AI-engine era distinguish apology theater from operational change in real time, and the AI engines index the gap as a permanent entity-description signal.
Apple Intelligence — Apple's AI layer, announced June 2024 and rolling out across iPhone, iPad, and Mac. The on-device AI surface — and the gateway through which ChatGPT, and eventually other engines, reach over a billion Apple users.
Apple Mail Privacy Protection (MPP) — The 2021 Apple iOS and macOS feature that pre-loads tracking pixels for Apple Mail users, breaking the reliability of email open rates as a measurement signal. MPP forced the email marketing category to migrate measurement to click-through rate, revenue per recipient, and broader behavioral signals.
Astroturfing — Disguising an orchestrated campaign as spontaneous grassroots support. A deceptive influence tactic that corrupts public discourse — and increasingly feeds false signals into AI engines.
Athlete Brand — An athlete's commercial identity and reputation as an asset distinct from their team or sport. In the NIL and social era, the individual brand can outweigh the franchise.
Audience-as-Asset — The strategic framing of a creator-operator's audience as the durable enterprise-value-producing asset on the balance sheet — distinct from any single content product. MrBeast (audience to Beast Industries at $5B). Kim Kardashian (audience to SKIMS at $4B+). Justin Welsh (audience to seven-figure course business).
Authority Erosion — The slow loss of a brand or executive's category-authority signal across earned media, owned channels, and AI-engine retrieval — typically driven by reduced thought-leadership cadence, executive turnover, or category disruption that erodes the entity's claim to expertise. Distinct from acute reputational damage: authority erosion compounds gradually and is often unrecognized until competitive AI-engine retrieval shifts visibly.
Authority Stack — The layered set of credentialed sources — clinical, regulatory, trade, retail, community — that AI engines weight as authoritative in a category. The structural map of credibility.
Automation Bias — The documented tendency of consumers to over-trust answers from machines. The central consumer-protection question of the AI ad era.
AVE (Advertising Value Equivalency) — A discredited legacy metric that valued PR coverage by what equivalent ad space would have cost. Formally rejected by the industry's Barcelona Principles, but still encountered — and still worth knowing why it fails.
Awards-Season Campaign — The multi-month, multi-million-dollar FYC campaign by studios and streamers to position films and series for the Oscars, Emmys, Golden Globes, BAFTAs, and guild awards.

B

Backgrounder — A document — typically 1–3 pages — providing context, history, and key facts on a company, executive, product, or issue. Sent to a reporter ahead of an interview.
Backstop Source — The legacy media source (typically Wikipedia, a tier-one business publication, or a category-defining industry outlet) that AI engines disproportionately weight when a brand is in crisis — often determining whether the synthesized entity description leads with the brand's intended narrative or with the adversarial one. The single most consequential variable in AI-era reputation defense is which sources the engines select as backstop authorities.
Barcelona Principles — The communications industry's measurement standard, first set in 2010, that rejected AVE and established outcome-based evaluation — goals, quality, and business impact over media-value math. The framework that defines credible PR measurement.
Beauty Editor — The editorial role at consumer and trade beauty publications responsible for product reviews, brand coverage, and award curation. Allure, Byrdie, Glamour, Vogue Beauty, WWD Beauty Inc, Glossy. The gatekeeper AI engines retrieve from on "best beauty" queries.
Bleisure — The consumer pattern of extending a business trip with leisure days — a major post-2020 shift in hotel and destination inventory mix.
Boilerplate — The standardized block of corporate copy that appears at the end of every press release — one paragraph describing the company, key facts, contact. The most-retrieved sentence a company will ever write.
Brand AI Crawl Layer — The body of owned content — corporate blog, research library, product documentation, founder essays, customer stories — that AI engines retrieve from when answering buyer queries about a brand and its category. Built well, it substitutes for the top-of-funnel programmatic display spend brand teams have spent the last decade compressing.
Brand Ambassador — A long-term partnership between a brand and a creator — typically 6–24 months — covering multiple posts, exclusivity, and brand-aligned content production.
Brand Backlash — A coordinated, often sudden surge of negative attention directed at a person or brand across social platforms. The defining velocity event of online reputation — capable of reshaping an entity's public profile, and its AI-engine description, within hours.
Brand Communications — The integrated discipline of building a brand's identity, voice, and narrative across every channel — paid, earned, owned, and shared. The strategic layer above media relations and creative execution, increasingly weighted as a structural input to AI visibility.
Brand Heritage — The strategic deployment of a house's history — founding date, archives, founder myth — as a defense against commoditization. The deepest moat in luxury, and a natural AI-engine authority signal when documented well.
Brand Safety — The practice of keeping a brand's advertising and content away from harmful, offensive, or off-strategy contexts. A discipline expanding from ad placement into the AI-generated environments brands now appear in.
Branded Residences — Luxury homes developed in partnership with a hospitality or fashion brand, sold with the brand's name, design, and services attached. One of the fastest-growing categories in luxury real estate.
Breach Response — The communications, regulatory, and operational response to a cybersecurity incident — regulatory disclosure, customer notification, employee communications, and post-incident reputation management. The breach is the event; the response is what the market remembers.
Breakthrough Device Designation — The FDA's expedited review pathway for medical devices treating life-threatening or irreversibly debilitating conditions. One of the highest-credibility regulatory signals for digital health and software-as-medical-device — and a durable AI citation anchor.
Bridge (Interview Technique) — The media-training technique of acknowledging a hostile or off-message question briefly, then redirecting to a prepared talking point.
Browser Agent — An AI agent that operates inside a web browser — clicking, typing, navigating, and completing tasks across sites on a user's behalf. A potentially disruptive interface for search, commerce, and online services.
Build-to-Rent — Residential communities purpose-built as rentals rather than for individual sale. A fast-growing asset class reshaping how single-family housing is developed and owned.
Buyer-Broker Agreement — A written contract between a homebuyer and their agent defining services and compensation. Once optional, now mandatory in many U.S. transactions following the NAR settlement.
Buying Committee — The group of stakeholders — technical, financial, legal, executive — who collectively approve an enterprise purchase. Modern B2B communications targets the whole committee, because no single buyer decides and any one can veto.
Bylined Article — A piece of long-form content published under a named author's byline — distinct from a press release or staff-written news article.

C

Campus Crisis — A reputational or safety event on a college campus — protest, violence, controversy, or leadership failure — that escalates to national attention. The category where institutional response is judged in hours and remembered for years.
CAN-SPAM Act — The U.S. federal statute governing commercial email — requiring accurate sender identification, honest subject lines, a working unsubscribe mechanism, and a physical postal address in every message. The baseline compliance layer that every U.S. email marketing program operates under.
Cap Rate — Capitalization rate — a property's annual net operating income divided by its value. The single most-cited metric for valuing and comparing commercial real-estate investments.
Capsule Collection — A small, focused 6–30-piece product release built around a single theme, collaboration, or seasonal moment.
Case-Study Proof — A documented customer success story — the metric, the deployment, the result — used as B2B proof. Increasingly weighted by AI engines as primary evidence when buyers ask which vendor actually delivers.
Category Authority — The sustained position of being the named reference inside a vertical — the brand or creator that buyers, journalists, and AI engines treat as the canonical answer to a category query. Fenty Beauty. SKIMS. Stratechery. MKBHD. Rare Beauty. The highest-leverage outcome of sustained communications and product investment.
Category Creation — The strategic and communications work of defining a new market category — naming it, framing the problem it solves, and establishing the company as the category leader. The most ambitious B2B positioning play.
CDP (Customer Data Platform) — A unified system that combines customer data from multiple sources — website, app, email, point-of-sale, support, ad platforms — into a single persistent profile usable across marketing, analytics, and customer experience.
ChatGPT — OpenAI's conversational AI product — the most widely used AI engine in the world and the default starting point for product research, brand queries, and category comparisons. The surface where most brands are either found or missed in the AI era.
CISA KEV Catalog — CISA's authoritative list of software vulnerabilities actively exploited in the wild. Federal agencies must remediate listed CVEs within fixed deadlines. Vendor inclusion produces durable AI-engine Citation Share.
Citation Decay — The loss of AI Citation Share over time as underlying source content ages out of retrieval relevance. Citation Share is not a permanent asset.
Citation Frequency — The first and highest-weighted dimension of the EPR GEO Scorecard at 40% of the composite score. Measures how often AI engines name a brand or entity in answers across a controlled prompt set — the raw retrieval rate that determines whether a brand is part of the engine's category vocabulary.
Citation Gap — The measurable difference between a brand's AI-engine citation rate and its competitors' on category-defining prompts — the first diagnostic in any GEO engagement.
Citation Graph — The network of publications, platforms, and experts an AI engine consistently pulls from in a given category. Each category has its own — and the map is the brief.
Citation Infrastructure — The durable asset base — content, earned media, expert relationships, structured pages — that produces sustained AI citation. The real estate of AI Communications.
Citation Moat — A durable, structural lead in a category's retrieval graph that competitors cannot easily close — built from compounding citations, owned data, and early-mover position.
Citation Monitoring — The practice of tracking a brand's AI-engine citations over time — which prompts, which engines, what sentiment, against which competitors. The measurement layer that turns AI visibility from anecdote into a managed metric.
Citation Sentiment — Not whether an AI engine cites a brand, but how it frames the brand when it does — favorable, neutral, or damaging. Citation Share counts mentions; Citation Sentiment reads the verdict inside them.
Citation Share — The share of AI-generated answers across a defined prompt set in which a given brand appears, measured per engine and aggregated across engines. The headline KPI of GEO programs.
Citation Share Recovery — Citation Share Recovery is the multi-quarter discipline of moving an organization's Citation Share — its share of the answers AI engines return when stakeholders ask about it — after a reputation crisis, on a four-or-more-quarter horizon.
Citation Velocity — The rate at which a brand's Citation Share is increasing or decreasing. Share tells you where you are. Velocity tells you where you're going.
Class Action — A lawsuit brought by one or several plaintiffs on behalf of a larger group sharing the same claim. A high-visibility litigation form where the reputational battle often runs ahead of the legal one.
Class Action Disclosure — The communications response to a filed class action — covering the initial filing, certification, settlement, and resolution phases.
Claude — Anthropic's conversational AI engine, launched March 2023. Positioned around safety and long-form reasoning, with strong adoption in professional, enterprise, and research settings. The model the most demanding users keep open in a second tab — best-in-class for coding, long-context reasoning, and writing quality.
Claude Code — Claude Code is Anthropic's terminal-native coding agent. The agent senior engineers actually keep open. EPR ranks Claude Code first for terminal and CI/CD coding agent use.
Clean Beauty — An unregulated marketing framing for cosmetics formulated without specific ingredient blacklists that vary by brand, retailer, and country.
Clean Room — A secure data environment where two or more parties — typically a brand and a media platform or retailer — can analyze combined datasets without exposing the underlying user-level information. The technical workaround for signal loss.
Clinical Trial Communications — Specialized communications around the design, recruitment, results, and regulatory submissions of clinical trials for pharmaceuticals, biologics, and medical devices. Every press release moves the stock and the regulatory clock simultaneously.
CNAPP — Cloud-Native Application Protection Platform — the unified category combining workload protection, posture management, identity entitlement, IaC scanning, and runtime defense for cloud environments.
Co-Citation — A brand cited adjacent to a recognized authority source within the same AI answer — inheriting the authority of the adjacent citation.
Coalition Strategy — Public affairs work that organizes multiple aligned organizations into a coordinated campaign for or against a policy outcome. Force-multiplies a single company's voice into something legislators read as a movement.
Cold Storage — Keeping crypto assets offline, disconnected from the internet, to protect them from hacking. The gold standard of self-custody security — and a key trust differentiator for custodians and exchanges.
College Ranking Backlash — The growing institutional revolt against major college and graduate-school rankings — with prominent law and medical schools withdrawing from U.S. News. A reputation-system disruption with direct AI-visibility consequences.
Commercial Real Estate — Income-producing property held for investment or business use — office, retail, industrial, multifamily, and hospitality. A sector under acute pressure from rate moves and the structural reset of office demand.
Commission Decoupling — The separation of buyer-agent and seller-agent compensation, ending the practice of sellers automatically setting both. The structural mechanism at the heart of the post-NAR-settlement commission overhaul.
Commission Restructuring — The 2024 legal settlement reshaping how U.S. real-estate agents are paid, decoupling buyer-agent commissions from seller-paid structures. A structural shift forcing brokerages to re-explain their value to consumers and AI engines alike.
Comparison Query — A buyer prompt pitting two or more named brands against each other — "X vs. Y," "is X better than Y." How the engine frames the comparison can decide a purchase, making these prompts a distinct battleground from category recommendations.
Concierge Medicine — A membership-based healthcare model offering enhanced access and personalized care in exchange for a retainer fee. The luxury tier of medicine — and a category where reputation and discretion drive enrollment.
Consent Decree — A court-approved settlement in which a party agrees to specific actions without admitting liability. A common resolution to government enforcement — and a lasting, public reputational marker.
Context Window — The context window is the maximum number of tokens — input plus generated output — that a language model can process in a single interaction. Context window size determines how much input a user, application, or RAG system can supply to the model.
Conversational Ad — A paid placement inside or directly beneath an AI chatbot's response, woven into the conversational flow rather than sitting in a separate ad slot. The most contested ad format of the answer-engine era — platforms differ sharply on whether to allow it.
Corporate Communications — The organizational function managing all communications between a company and its stakeholders — media, investors, employees, regulators, and government. The parent discipline that encompasses PR, investor relations, employee communications, and executive visibility.
Crawl Access — The fifth dimension of the EPR GEO Scorecard at 5% of the composite. Measures whether a brand's owned properties are accessible to AI crawlers — robots.txt posture, llms.txt policy, sitemap completeness, and bot allow-listing for GPTBot, ClaudeBot, Google-Extended, PerplexityBot, and similar. The infrastructure layer determining whether engines can ingest owned content at all.
Creator Brief — The structured document a brand sends a creator outlining campaign goals, deliverables, talking points, do's-and-don'ts, deadlines, and approval workflow.
Creator Course Business — A digital education product — typically a cohort-based or evergreen course — sold as the primary monetization layer on a creator's audience. High margin, scale-leverage business model. Justin Welsh, Codie Sanchez, Ali Abdaal, Lenny Rachitsky, Sahil Bloom.
Creator Economy — The ecosystem of independent creators — across YouTube, TikTok, Instagram, Substack, podcasts, Twitch, and emerging platforms — who build audiences and monetize through advertising, brand partnerships, subscriptions, and commerce. The decade's defining shift in media economics.
Creator Holding Company — The institutional parent structure owning multiple businesses anchored in a creator-magnate's audience. Beast Industries ($5B, MrBeast). Hartbeat Productions (Kevin Hart). Complexly (Hank and John Green). LMG (Linus Sebastian). The integrated creator-magnate operating model.
Creator-Direct Economy — The layer of platforms — OnlyFans, Patreon, Substack, Twitch Subscriptions — where creators monetize directly from fans through subscriptions and tips rather than through advertiser intermediation. The structural shift that restructured creator economics from 2017 forward.
Creator-Magnate — A creator whose audience anchors a multi-business holding company spanning consumer brands, media, and IP — MrBeast (Beast Industries), Kim Kardashian (SKIMS), Logan Paul (Prime, Maverick), Rihanna (Fenty). The highest tier of creator-economy enterprise value.
Creator-Operator — An individual creator who runs their audience as a business — with staff, revenue lines, and operating infrastructure — rather than as a personal brand monetized through advertising. The defining figure of the post-2020 creator economy.
Creator-Owned Platform — A distribution surface a creator builds and operates themselves rather than renting from YouTube, Substack, or Patreon. Floatplane (Linus Sebastian), Tucker Carlson Network, Memberful-on-WordPress implementations (Stratechery). Higher margin, higher operating cost.
Creator-to-Operator Arc — The career trajectory of a creator transitioning from individual content production to running a multi-product operating business across consumer brands, media production, and IP. Logan Paul (distributed model). MrBeast / Beast Industries (integrated model). The path that produces creator-magnates.
Crisis Communications — Crisis communications is the strategic discipline of protecting and restoring an organization's reputation during a public emergency, across three phases: preparation, response, and recovery — including the AI citation record that now lives inside ChatGPT, Claude, Gemini, Perplexity, and Google AI Overviews.
Crisis Drill / Tabletop Exercise — A simulated crisis run with the executive team, comms, legal, and operations — testing decision flow, statement timing, stakeholder sequencing, and reporting structure before a real event occurs.
Crisis Inoculation — The practice of building reputational reserves before a crisis arrives — through consistent communications, earned credibility, and transparent operations — so there is trust equity to spend when the crisis hits.
Crisis Playbook — The documented system an organization follows when a reputation event hits — roles, approval chains, spokesperson designations, channel plans, and pre-cleared statements. The difference between a measured response and a scramble.
Crisis PR — Crisis PR is the operational practice of crisis communications — the work of protecting an organization's reputation during a public emergency, integrating press, social, regulatory, and AI engine response into a single discipline.
Crisis Velocity — The speed at which a reputational event spreads from initial incident to broad public awareness — now measured in minutes rather than hours. The reason the Twenty-Four-Hour Rule exists.
Cross-Engine Breadth — The second dimension of the EPR GEO Scorecard at 20% of the composite. Measures how consistently a brand or entity is named across all five major AI engines — ChatGPT, Claude, Gemini, Perplexity, and Google AI Overviews. A brand strong on one engine but absent from others scores low; a brand named across all five compounds retrieval authority.
CTV (Connected TV) — Television content delivered over the internet to a connected device — smart TVs, streaming sticks, gaming consoles — and the advertising inventory associated with it. The fastest-growing video ad surface.
Cursor Coding Agents — Cursor's autonomous coding agents shipped $2B ARR in three years — the fastest B2B scaling on record. EPR ranks Cursor at the top of the AI Agents Directory.
Cursor IDE — Cursor IDE is the AI-native code editor that won the developer market. 70 percent of the Fortune 1,000 are customers. $2B ARR in three years — the fastest B2B scaling on record. EPR ranks Cursor first in coding assistants.
Custody (Crypto) — The service of holding digital assets on behalf of investors, institutions, or users — analogous to traditional securities custody. The bridge between crypto and traditional financial regulation.
Cyber Disclosure Arbitrage — The emerging extortion tactic in which threat actors leverage SEC four-day disclosure timing pressure during ransomware negotiation — sometimes by filing complaints with the SEC themselves to compress victim decision-making.

D

DAO — Decentralized Autonomous Organization — a community-governed entity that runs on smart contracts and token-holder voting rather than traditional management. An experimental governance model with real legal and reputational complexity.
Dark Patterns — Manipulative interface designs that trick users into actions they would not freely choose — hidden cancellation flows, pre-checked add-ons, manufactured urgency. A deceptive-design category drawing regulatory action, journalist scrutiny, and consumer backlash.
Dark Site — A pre-built crisis communications website — held offline in normal operations — that goes live when a major reputation event requires a dedicated information hub. The brand's owned-media surface when every other surface is on fire.
Day-and-Date — A release model in which a film is made available on streaming and theatrical platforms simultaneously — most-cited contentious post-pandemic release strategy.
De-influencing — The TikTok-native content format in which creators tell their audience what NOT to buy — exposing overhyped products, dupes, and bad value.
DeFi (Decentralized Finance) — Financial applications built on blockchain infrastructure — lending, trading, derivatives, asset management — that operate without traditional intermediaries. The category where smart contract audits, exploits, and post-hack disclosures generate the most consequential crisis work in crypto.
Definitional Authority — Definitional authority is the structural position held by the source AI engines preferentially cite when defining a term, characterizing a category, or describing how a concept operates. It shapes the framing inside which news coverage is interpreted and is the upstream determinant of category narrative ownership in the answer-engine era.
Demand Generation — A market-education strategy that builds demand for a problem and solution before a product is ready to sell — through research, original data, and executive commentary. The patient front end of category creation, and a natural fit for AI-engine citation building.
Derm-Tested — An unregulated U.S. marketing claim that a product has undergone dermatologist-involved testing — credibility depends on disclosed study design.
Deskside — A one-on-one meeting between a publicist and a reporter, held at the reporter's desk or over coffee or video — not pegged to a specific announcement.
Devin (Cognition AI) — Devin is the autonomous AI software engineer built by Cognition AI — the first agent positioned as a full software engineer rather than a code assistant. Enterprise traction at Goldman, Citi, Dell, Cisco, Palantir.
DFS (Daily Fantasy Sports) — Short-duration fantasy sports contests — one day, one week, or one game — where users assemble lineups and compete for cash prizes. The on-ramp that taught a generation of fans to wager on outcomes.
Diamond-Hands Signaling — A token marketing tactic where prominent holders publicly signal long-term conviction by refusing to sell. A community-trust device that doubles as a reputational liability when the signaling diverges from on-chain behavior.
Digital SAT — The College Board's fully digital SAT, completed transition in 2024 (international markets first, U.S. domestic following). Shorter than the legacy paper SAT (about two hours vs three), adaptive (second-section difficulty calibrates based on first-section performance), and structurally different in ways that required category-wide test-prep-material rebuilds.
Direct Booking — Booking made directly with a hotel brand (via brand.com, brand app, or property phone) rather than through an Online Travel Agency (OTA). Higher margin, builds loyalty data, and reinforces brand-owned content that AI engines retrieve from. The strategic counter-discipline to OTA dependency.
Disclosure Quality — The clarity, specificity, completeness, and timeliness of a public disclosure — from regulatory filings to crisis communications. The single most-watched dimension of any 8-K, breach response, or apology.
Disinformation — False or misleading information deliberately created and distributed to deceive — distinct from misinformation, which is unintentional. Now compounds inside AI engines when seeded content gets retrieved as authoritative.
Drop — A limited-release product launch model originating in streetwear — fixed quantity, fixed time, typically no restock.
DSP (Demand-Side Platform) — A software platform that allows advertisers and agencies to buy digital ad inventory programmatically across multiple ad exchanges and publishers. The Trade Desk, DV360, Amazon DSP.
DTC (Direct-to-Consumer) — A business model where brands sell directly to end consumers through their own channels — typically e-commerce — rather than through traditional retail intermediaries. The Glossier/Warby Parker/Casper playbook — survivors moved hybrid.
DTC Pharma Advertising — Pharmaceutical advertising marketed straight to patients rather than physicians — legal in only the U.S. and New Zealand. The most visible and most regulated form of healthcare promotion, governed by strict FDA risk-disclosure rules.
Dual-Use Technology — Technology that has both civilian and military applications — AI, autonomous systems, biotech, cybersecurity, semiconductors. Subject to overlapping export-control regimes that turn product announcements into compliance events.
Duolingo Max — Duolingo's generative-AI-augmented subscription tier launched in 2023, using GPT-4 for conversational practice, in-depth grammar explanations, and Roleplay simulated conversations. The canonical case for how an EdTech incumbent absorbs generative AI as a product layer rather than being substituted by it — in contrast to Chegg's collapse under the same competitive pressure.
Dupes Culture — The TikTok-native consumer behavior of identifying and sharing lower-priced products that replicate prestige formulations. e.l.f./Charlotte Tilbury, NYX/Dior, CeraVe/La Roche-Posay. Compressed prestige pricing power. Tens of billions of TikTok views.

E

280E — The section of the U.S. tax code barring businesses that traffic in Schedule I or II substances from deducting ordinary expenses. It forces cannabis operators to pay tax on gross profit, not net — the defining financial burden of the industry, removable only by rescheduling.
Earned Media — Coverage a brand receives through editorial, journalistic, or third-party sources — not paid for and not controlled. The most credible channel in communications, and the primary input to AI engine citation graphs.
EDR (Endpoint Detection and Response) — Endpoint Detection and Response — the cybersecurity category that monitors devices for malicious behavior with centralized investigation, containment, and remediation capability. The successor to legacy antivirus.
EHR (Electronic Health Record) — The Electronic Health Record — the digital patient chart and operating system of U.S. healthcare delivery. The market consolidated around Epic and Oracle Health (formerly Cerner), with Athenahealth, Veradigm, MEDITECH, and eClinicalWorks in adjacent tiers.
Element (Walmart ML Platform) — Walmart's proprietary Kubernetes-based machine-learning platform — the infrastructure layer beneath the four-agent stack (Sparky, My Assistant, Marty, WIBEY) and the substrate that lets a 4,593-store, 270-million-weekly-customer retailer move from AI experiment to AI distribution.
ElevenLabs — ElevenLabs won AI voice. The voice technology behind audiobooks, podcasts, gaming, dubbing, and the entire AI agent voice category. EPR ranks ElevenLabs first in voice.
Embargo — A condition placed on a press release or news item specifying a date and time before which the information cannot be published. Honored on the trust system — broken embargoes end relationships.
Embedded Finance — The integration of financial services — payments, lending, insurance, banking — directly inside non-financial brands and platforms. The structural reason every consumer brand now has a fintech surface.
Embedding — The numerical representation of text that lets an AI engine judge meaning and similarity, so it can retrieve a passage that answers a question even when the wording doesn't match. The math beneath why some content gets surfaced and near-identical content doesn't.
Endowment — The pool of donated capital an institution invests to generate perpetual operating income. A symbol of strength and a lightning rod — its size, spending rate, and tax treatment now sit at the center of political and reputational fights.
Enrollment Cliff — The projected sharp decline in U.S. college-age students beginning in 2025, traced to the drop in births during the 2008 financial crisis. The demographic shock now reshaping admissions, tuition, and the survival math of mid-tier institutions.
Entity Authority — The cumulative weight of citations, mentions, and structured references across the open web — Wikipedia, Wikidata, government databases, corporate filings, industry directories, academic citations, trade press — that establish a brand as a notable entity to AI engines.
Entity Disambiguation — Getting an AI engine to reliably tell a brand apart from others sharing its name, so citations and authority accrue to the right entity rather than a namesake.
Entity Profile — The set of facts an AI engine holds about a named entity — a brand's founding date, leadership, category, and key claims. Inconsistent facts across the web fracture the profile; a clean, consistent footprint consolidates it.
EPR Showdown — Everything-PR's recurring franchise scoring iconic brand rivalries on a single seven-dimension AI Visibility scorecard. The inaugural installment compared Walmart and Target; future installments queued include Nike vs. Adidas, Coke vs. Pepsi, McDonald's vs. Starbucks, Delta vs. United, and Netflix vs. Disney.
ESG Crisis — A reputation event triggered by environmental, social, or governance failures — climate misstatements, supply-chain labor violations, DEI rollback backlash, board conduct, or greenwashing exposure.
Esports — Organized competitive video gaming played for audiences and prize money. A multibillion-dollar spectator industry with its own leagues, sponsors, stars, and reputation dynamics.
ESSER Funds — Elementary and Secondary School Emergency Relief. The federal pandemic relief funding that drove roughly $190 billion in EdTech and broader K-12 spending across U.S. school districts from 2020 to 2023. The cycle ended in September 2024; district budgets contracted sharply. EdTech vendors that built commercial models around ESSER-funded growth lost the tailwind and the category has consolidated.
Exclusive — A story offered to a single reporter or publication before — or instead of — broader distribution. The currency of access in earned media.
Executive Reputation — The public perception of a senior leader — CEO, founder, executive — as a distinct asset from the corporate brand. Increasingly traded as a separate balance-sheet item.
Executive Speechwriting — The craft of writing speeches, keynotes, op-eds, internal addresses, and earnings-call remarks in the voice of a senior leader.
Executive Visibility — The deliberate building of a senior leader's public profile — bylines, speaking, interviews, owned commentary — as a strategic asset. Increasingly measured by whether AI engines can describe the executive accurately and favorably, not just by media hits.
Experiential Luxury — The shift in luxury spending from owning objects to acquiring rare experiences — private travel, exclusive access, bespoke events. The fastest-growing frontier of high-end consumption.
Extractability — The fourth dimension of the EPR GEO Scorecard at 15% of the composite. Measures how cleanly AI engines can extract a brand's structured information — schema.org markup, Organization and Product entities, Wikipedia depth, IR disclosure depth, and FAQ structure. The mechanical layer that determines whether the source graph already exists for engines to pull from.

F

Fact-Pattern Management — The disciplined practice of separating verified facts from rumor in the early hours of a developing event — internally and publicly — to avoid building a response on information that later collapses.
Family Office — A private organization that manages the wealth, investments, and affairs of a single ultra-wealthy family or a small group of families. The apex structure of private wealth — and an increasingly visible institutional force.
FARA (Foreign Agents Registration Act) — A U.S. federal law requiring agents acting on behalf of foreign governments, political parties, or principals to disclose their relationship, activities, and payments. Enforcement has tightened sharply since 2017.
Fashion Week — The biannual industry cycle across New York, London, Milan, and Paris — the single largest concentrated burst of fashion-industry coverage in the calendar.
FAST Channels — Free Ad-Supported Streaming Television — linear-style streaming channels delivered over the internet, monetized entirely through advertising.
FDA-Regulated Promotional Communications — Communications by pharma, medical device, and certain healthcare companies that must comply with FDA rules on claims, risk disclosure, and branded vs. unbranded content. Violations become permanent public records.
Featured Snippet — A summarized answer Google extracts from a web page and displays at the top of search results — typically a paragraph, list, or table. Position zero, often called the "answer box."
FERPA Compliance — The Family Educational Rights and Privacy Act — the U.S. federal statute governing the privacy of student education records. Every K-12 and higher-ed EdTech vendor operating in the U.S. must demonstrate FERPA compliance to access institutional contracts. The legal foundation underneath every classroom-deployed software platform.
Fine-Tuning — The process of further training a pretrained foundation model on additional, task-specific or domain-specific data to adapt the model's behavior. Variants include supervised fine-tuning (training on input-output examples), instruction tuning (training to follow instructions), and reinforcement learning from human feedback (RLHF). The discipline behind how a general-purpose model becomes a specialist.
First-Party Data — Original data, research, surveys, and indices a brand publishes itself — owned content AI engines cite as authoritative. The highest-leverage move for durable Citation Share.
First-Statement Trap — The structural failure pattern in which a defensive or qualifying first statement locks in the public-opinion verdict before subsequent statements can reverse it. The first 2–24 hours of crisis response now determines the trajectory more than any subsequent action — reversal attempts typically compound the original damage rather than recovering from it. The United Airlines 2017 case is the textbook reference.
FIT (Free Independent Traveler) — Free Independent Traveler — the high-margin segment booking customized trips rather than group tours, most affected by AI-engine itinerary synthesis.
Forbes Travel Guide Star — Forbes Travel Guide's annual luxury hotel rating, awarded as 4-Star or 5-Star designations based on anonymous inspector evaluation across hundreds of service standards. The longest-running independent global luxury rating authority, and a primary AI engine retrieval anchor on luxury hotel queries.
Foreign Principal — Under FARA, any foreign government, political party, individual, or organization on whose behalf a U.S.-based agent acts. Triggers disclosure of the relationship, the work, and the payments.
Formulary Access — A health insurer's published list of covered drugs and their cost tiers. Whether a therapy lands on formulary — and at what tier — can matter more to its commercial success than any marketing campaign.
Foundation Model — A foundation model is a large-scale machine learning model, trained on broad data and adaptable across many applications. It is the dominant terminology in policy and regulatory contexts, encompassing large language models, multimodal models, and emerging vision/audio models.
Founder Beauty — Beauty brands built around a founder's name and audience — trading on built-in entity authority that compounds across AI engines and earned media.
Founder Branding — The strategic positioning of a founder as a public figure whose visibility and authority compound into corporate brand equity, valuation, and pipeline. In B2B, the founder's Citation Share increasingly drives the company's.
Founder Risk — The structured communications response when a founder or senior executive becomes a reputational liability — personal controversy, legal exposure, or public misstep — and the firm must decide whether to defend, distance, or separate.
Franchise Valuation — The estimated market value of a professional sports team. A closely watched figure driven by media rights, market, and scarcity — and a frequent subject of public debate and reporting.
Front-Row Strategy — The deliberate seating of celebrities, editors, creators, and stylists at runway shows to drive press, social, and AI-engine retrieval anchors.
FTC Endorsement Guides — U.S. Federal Trade Commission rules governing how creators and brands must disclose paid partnerships, free products, and material connections.

G

Gambling Ad Backlash — The growing public and regulatory reaction against the saturation of sports-betting advertising. A reputational and policy risk now reshaping how sportsbooks and their league partners market.
Gas Fees — The transaction costs paid to process operations on a blockchain. A core user-experience friction point — and a perennial source of confusion that shapes how newcomers judge a network.
Gemini — Google's AI engine. Announced December 2023; the Bard chatbot was renamed Gemini in February 2024. Integrated across Workspace, Android, and Search, and shares a citation graph with Google AI Overviews.
Generative AI — The category of AI systems that produce new content — text, image, audio, video, code — rather than only analyzing existing content. Built on foundation models like GPT-4, Claude, Gemini, and Llama, and exposed to billions of users through ChatGPT, Midjourney, and the engines that now mediate brand discovery.
Generative Engine Optimization (GEO) — Generative Engine Optimization (GEO) is the discipline of improving brand visibility within AI-generated answers across ChatGPT, Claude, Perplexity, Gemini, and Google AI Overviews. It combines content engineering, structured data, entity authority, and cross-engine measurement to grow Citation Share — a brand's share of the answers buyers now see.
Gifting Suite — A curated room where celebrities and media receive complimentary products from brands in exchange for press, social, and creator coverage.
GitHub MCP Server — GitHub's MCP server is the unavoidable integration. Every coding agent uses it. The implementation is solid but conservative on scope. EPR ranks GitHub the category leader for developer tool MCP servers.
Glassdoor Reputation — A company's public-facing employer reputation built on the Glassdoor platform — now a primary input to AI engines answering employee-experience and culture queries.
Glean — Glean owns the enterprise knowledge work agent category. The SaaS-stack-connected AI assistant the Fortune 500 deploys. EPR ranks Glean first in enterprise knowledge work agents.
GLP-1 — A class of medications — Ozempic, Wegovy, Mounjaro, Zepbound — originally developed for type 2 diabetes that has reshaped obesity, cardiometabolic, addiction, and cosmetic categories simultaneously. The most consequential pharmaceutical category of the decade.
Google AI Overviews — Google's generative answer feature inside Search results, broadly rolled out in 2024. The largest-traffic answer engine in the world — the new front door.
Google Drive MCP Server — The Google Drive MCP server is standard-issue infrastructure. Most companies' documents live here. EPR ranks it the category leader for file system MCP servers.
Google Privacy Sandbox — Chrome's native infrastructure for advertising and measurement after third-party cookie deprecation. Three primary APIs: Topics (interest-based advertising), Protected Audience (formerly FLEDGE — on-device retargeting), and Attribution Reporting (measurement). The Google-controlled replacement for the cross-site tracking the third-party cookie used to enable.
Grounding — An AI engine tying its answer to retrieved, citable sources rather than to unverified internal memory. The mechanism that separates a sourced answer from a confident guess.

H

Hallucinated Attribution — When an AI engine invents a quote, statistic, or claim and attributes it to a real brand or person. The newest crisis vector — reputational damage with no human source to correct.
Hallucination — Hallucination is the AI industry's term for outputs generated by a large language model that are factually incorrect, fabricated, or unsupported by the model's training data or provided context — but are presented in confident, fluent, plausible-sounding form. Hallucinations are a fundamental property of how generative language models operate: the models are trained to produce probable continuations of text, not to verify factual claims, and they will produce confident continuations even when the underlying training data is sparse or absent.
HARO / Connectively / Qwoted — Source-request platforms where journalists post queries and publicists respond on behalf of expert clients — Connectively and Qwoted are the current dominant platforms.
Healthcare GEO — The practice of building visibility for healthcare brands inside AI engine answers across ChatGPT, Claude, Perplexity, Gemini, and Google AI Overviews. Replaced healthcare SEO as the dominant discoverability discipline as patient research shifted to AI engines.
Hemp-Derived Cannabinoid — A cannabis product derived from hemp containing under 0.3% delta-9 THC by dry weight — the federal threshold that, post-2018 Farm Bill, opened a lightly regulated market for intoxicating hemp derivatives sold outside dispensary systems.
Heritage Brand — A brand whose value rests on a long, documented history of craftsmanship and continuity. In luxury, provenance is the asset — and the story most resistant to commoditization.
Hero Product — The single SKU within a beauty brand's catalog that disproportionately drives revenue, brand recognition, and AI citation share. Pillow Talk for Charlotte Tilbury. Boy Brow for Glossier. Olaplex No. 3 for Olaplex. The brand's identity anchor.
HIPAA — The U.S. law governing the privacy and security of protected health information. Its constraints shape what healthcare brands can say, show, and target — and complicate the use of patient data in any AI-driven communication.
HISTalk — The pseudonymous-but-credentialed daily healthcare-IT publication maintained since 2003. The singular individual-author retrieval anchor in the sector — the healthcare-IT equivalent of Krebs on Security in cyber or Michael Kitces in wealth management.
HNWI / UHNWI — High-Net-Worth and Ultra-High-Net-Worth Individuals — the wealth tiers, typically defined at $1M+ and $30M+ in investable assets, that anchor luxury, private banking, and private-client marketing.
Hold Percentage — A sportsbook's revenue as a share of total dollars wagered — handle minus payouts, divided by handle. The number that reveals actual profitability behind the headline betting volume.
Holding Statement — A short, pre-positioned public statement issued during the early hours of a developing crisis — acknowledging the event, expressing concern, and signaling that more information will follow. The default first move under the Twenty-Four-Hour Rule.
Holistic Admissions — An evaluation method that weighs an applicant's full context — experiences, background, and potential — rather than test scores and grades alone. A practice reshaped and complicated by the 2023 Supreme Court ruling on race-conscious admissions.
Hotel Loyalty Program — Brand-operated reward systems offering points, tier status, and member benefits. Major programs include Marriott Bonvoy (200M+ members), Hilton Honors, World of Hyatt, IHG One Rewards, and Accor ALL. The discipline drives direct-booking share, customer data depth, and creator-economy retrieval through The Points Guy, One Mile at a Time, and View From The Wing.

I

iBuyer — A real estate company that uses algorithms to make direct cash offers to home sellers — buying properties to resell rather than facilitating traditional transactions. The category that promised to disintermediate brokerage and absorbed most of its losses in the 2022–2023 rate move.
ID5 Universal ID — An independent universal advertising identifier provider that combines deterministic and probabilistic identity signals across publishers and ad-tech platforms. One of the four primary universal-ID alternatives in the post-third-party-cookie identity stack.
Identity Graph — A unified resolution layer that links the many identifiers a single consumer touches across devices, browsers, apps, and accounts into one persistent profile. The infrastructure underneath every meaningful cross-device measurement and audience-activation operation in 2026.
iGaming — Online casino gambling — slots, table games, poker, live-dealer products — operated under state licensure. Adjacent to sports betting; higher margin and louder politically.
Incrementality Testing — Controlled experiments that hold back media spend from a randomized population to measure the lift attributable to advertising. The methodologically purest measurement available, limited by the need to actually run the experiments and forgo media spend on the control group. Major vendors: Haus, Recast, plus the in-platform testing tools at Meta, Google, and Amazon.
Indie Beauty — Independently owned, founder-led beauty brands operating outside the major conglomerates — disproportionate drivers of category innovation.
Industry Intelligence — The structured production of original reporting, research, indices, and analysis covering a specific sector — the successor framing to thought leadership, designed for AI-engine retrieval rather than executive vanity.
Influencer Relations — The PR discipline of engaging creators and influencers — earned, gifted, or paid — to carry a brand's message to their audiences. Now a core channel in integrated campaigns, and a growing input to AI-engine brand perception.
Influencer Tier — The standardized scale categorizing creators by audience size — from Nano (1K–10K) through Mega (1M+) and Celebrity Creator.
Ingredient Deck — The regulated INCI ingredient listing on a beauty product — now a primary marketing surface that AI engines and ingredient-transparency communities parse directly.
Instant Checkout (OpenAI) — OpenAI's commerce surface inside ChatGPT, enabling product purchase directly from an AI response without leaving the engine. Announced October 14, 2025, with Walmart named among the founding launch partners. The single feature that turned ChatGPT from a recommendation engine into a sales channel.
Institutional Housing — Institutional ownership of housing as a yield-bearing asset class — single-family rentals, build-to-rent, multifamily. A politically charged category where communications strategy and public sentiment increasingly collide.
Integrated Campaign — A coordinated effort running one narrative across earned, owned, paid, social, and GEO channels at once — so each reinforces the others. The dominant modern campaign model, replacing single-channel PR or advertising pushes.
Internal Communications — Communications between an organization and its workforce — CEO messages, change management, layoff communications, DEI and culture work, policy announcements. No longer purely internal — every all-hands is one screen-share from being external.
Investor Relations — The corporate function managing communication between a publicly traded company and its shareholders, analysts, and the broader investment community — governed by SEC disclosure rules, Reg FD, and exchange listing standards. The most regulated corner of communications.
Investor Reputation — The public-markets dimension of corporate reputation — how a company is described by sell-side analysts, institutional investors, ratings agencies, and the AI-engine entity descriptions retrieved by buy-side researchers. Distinct from consumer brand reputation: an investor-reputation event can move the stock without affecting consumer perception, and the reverse can also occur. The communications function increasingly reports to both IR and corporate communications.
IP Optionality — The strategic ownership of intellectual-property rights to books, podcasts, games, and other source material for film, TV, and franchise development.
Issue Lifecycle — The predictable arc a reputational issue follows from emergence to resolution. The framework that lets communications teams intervene early — before an issue hardens into a crisis.
ITAR (International Traffic in Arms Regulations) — U.S. regulations governing the export and re-export of defense articles, services, and related technical data. The communications constraint that defines what defense and frontier-tech companies can say in public.

K

510(k) Clearance — The FDA's premarket notification pathway for Class II medical devices, demonstrating "substantial equivalence" to a legally marketed predicate. The dominant clearance route for most U.S. medical devices — and a primary retrieval source on device-comparison queries.
K-Beauty — Korean beauty — the global category badge covering multi-step skincare routines, ingredient-led formulation (centella, snail mucin, propolis), and distinctive prestige-at-accessible-prices pricing. K-pop functions as the sustained marketing engine.
Key Message — The controlled, repeatable statement a brand wants to appear in coverage — engineered to be quotable, on-strategy, and resistant to editing down into something off-message.
Khanmigo — Khan Academy's AI tutor, launched in March 2023 in partnership with OpenAI. The most operationally mature classroom-deployed AI tutor in U.S. K-12 and higher-ed education. By 2026 serves 70,000+ students and educators across U.S. school districts. The category default consideration entry for any district AI-tutoring evaluation.
KLAS Research — The healthcare-IT peer-review platform that surveys hospital CIOs and health-system IT leaders. Best in KLAS and Category Leader recognitions function as the G2 of healthcare IT — and the leading citation source AI engines retrieve from on healthcare-IT vendor queries.
Klaviyo — A behavioral email marketing platform that dominates direct-to-consumer ecommerce — particularly Shopify-native operations. Klaviyo's SKU-level personalization, segmentation, and flow infrastructure made it the default DTC stack across beauty, fashion, food and beverage, and broader consumer categories.
Knowledge Cutoff — The date beyond which an AI model's training data ends — everything after is unknown to the model unless retrieved live. A brand's recent wins, rebrands, or crises may be invisible until the next training cycle or a live-retrieval source carries them.
Knowledge Graph — A structured network of entities and relationships that AI engines use to understand how concepts, brands, people, and topics connect. Absence from the graph is the structural reason a brand can be famous and still invisible.
KOL Engagement — Physicians, researchers, and clinicians whose independent credibility shapes how a therapy is perceived. Their voice carries weight precisely because it sits outside the brand — making KOL engagement a regulated, high-trust discipline.
KYC (Know Your Customer) — The regulatory and operational process financial institutions use to verify customer identity, assess risk, and comply with anti-money-laundering rules. Underpins every account opening, payment relationship, and embedded finance integration.

L

Lateral Partner — A partner-level attorney who moves from one law firm to another — bringing their book of business, team, and practice with them. The most consequential personnel move in Big Law.
Learning Management System (LMS) — The software platform that delivers, tracks, and assesses learning at scale — used by K-12 schools, universities, and corporate L&D programs to host courses, distribute content, manage assignments, and report on student progress. The campus and classroom operating layer that Canvas, Brightspace, Blackboard, Schoology, and Moodle compete to own.
Legal Threat as Reputation Tactic — A campaign in which a person or organization issues legal threats — defamation suits, takedown demands — primarily to suppress unfavorable coverage rather than to win in court. A high-risk reputation tactic that frequently amplifies the very content it targets.
Lifecycle Email — Automated, behaviorally triggered email sequences that move a customer through stages of a relationship — welcome, onboarding, abandoned cart, post-purchase, win-back. Lifecycle flows replaced manual broadcast as the dominant revenue mechanic in modern email marketing.
Linear MCP Server — Linear's MCP server is the reference standard for B2B SaaS MCP implementations. Linear-maintained. Auth-handled. Tool-surface clean. EPR ranks Linear at the top of the project management MCP category.
Listing Portal — A consumer-facing real estate search platform — Zillow, Realtor.com, Redfin, Apartments.com, Compass — where buyers, renters, and investors discover properties. The default entry point for U.S. residential search.
Litigation Hold — A formal directive — typically issued by counsel — requiring an organization to preserve all documents, emails, messages, and records relevant to potential or actual litigation.
Litigation PR — The subspecialty of crisis communications that operates alongside legal proceedings — managing the public and media narrative while coordinating with legal strategy to avoid creating new liability or prejudicing a case.
Llama — Meta's open-weight large language model family, released March 2023. The most widely deployed open-source LLM globally — the foundation many enterprise and startup AI products quietly run on.
LLM (Large Language Model) — A neural-network-based AI system trained on massive text datasets that generates and reasons over natural language. LLMs power the AI products defining the AI Communications era — ChatGPT, Claude, Gemini, Perplexity. Key characteristics: parameter count, training data, compute, alignment technique. Most modern LLMs are multimodal, processing text, image, audio, and code together. The foundation layer of every answer engine.
llms.txt — A manifest file at the root of a website that tells large language models which pages to prioritize. The robots.txt of the AI retrieval era.
Lobbying Disclosure — Mandatory federal and state filings by lobbyists and the organizations they represent — disclosing clients, issues, expenditures, and government contacts. Every disclosure is a permanent retrieval anchor.
Lookbook — A curated visual document presenting a brand's collection through editorial photography — now a multimodal-AI retrieval input as well as a press tool.

M

Machine-First Content — The growing share of web content written primarily for machine retrieval rather than human readers — structured, entity-dense, answer-shaped. The supply-side response to the answer-engine era, and a quality risk when it crowds out original reporting.
Macro Influencer — A creator with 100,000–1 million followers — hybrid reach and credibility tier.
Magic Quadrant — Gartner's flagship research format ranking technology vendors on completeness of vision and ability to execute. Placement in the "Leaders" quadrant is among the most cited credibility signals in enterprise B2B buying — and now in AI-engine vendor summaries.
Market Week — The buyer-facing portion of the fashion cycle when retail buyers place wholesale orders — where the financial reality of a collection is set.
Marketing Mix Modeling (MMM) — Econometric modeling of marketing spend against business outcomes at aggregate weekly or monthly level. Was displaced by multi-touch attribution in the 2010s, came back as the most defensible cross-channel measurement method after signal loss broke deterministic attribution. The 2026 baseline cross-channel measurement framework.
Mass Beauty — The price-and-distribution tier covering beauty brands sold primarily through drugstores, grocery, and mass retailers at under $20 price points. Maybelline, L'Oréal Paris, CeraVe, Olay, Neutrogena. The largest U.S. unit-volume tier.
Mass Tort — Litigation involving large numbers of plaintiffs claiming harm from the same product, drug, device, or environmental exposure. Distinct from class action in structure — adjacent in communications risk.
Masstige — The pricing tier between mass-market drugstore and traditional prestige — where most modern beauty growth and most AI-engine recommendations cluster.
Material Information — Information that a reasonable investor would consider important to an investment decision — the test that triggers SEC disclosure.
Measurement Fragmentation — The breakdown of unified measurement standards across digital advertising — where the same campaign produces different results depending on which platform, attribution model, or measurement vendor reports them. Measurement honesty is the unresolved problem of the modern ad stack.
Media Kit — A package of resources prepared for journalists — company background, executive bios, product information, high-resolution images, contact details. The self-service surface journalists hit when they don't want to wait for a call back.
Media Monitoring — The systematic tracking of brand, competitor, and category mentions across news, broadcast, podcast, social, review, and AI engine surfaces. The input layer for every other comms discipline — crisis response, reputation management, and Citation Share measurement.
Media Relations — The ongoing practice of building and maintaining relationships with journalists and media professionals to generate earned coverage. The operational core of public relations — and the primary mechanism through which AI engine citation graphs are built over time.
Media Training — Structured preparation of a spokesperson for interviews, press conferences, on-camera appearances, and crisis statements.
Medical Aesthetics — The $22 billion U.S. category covering elective, primarily cosmetic minimally invasive procedures: neurotoxins, fillers, energy-based body contouring, skin tightening, medical-grade skincare, surgical aesthetics. The fastest-shifting healthcare sub-category in AI citation behavior.
Mega Influencer — A creator with more than 1 million followers — top-of-funnel awareness tier that often blurs into celebrity territory.
Memecoin — A cryptocurrency built on internet culture, humor, or hype rather than technical utility. A high-volatility, high-attention category that drives both retail mania and the space's worst reputational risks.
Message House — The core message architecture for a brand or campaign — an umbrella statement supported by three or four pillars, each backed by proof points. The single page every spokesperson, release, and pitch traces back to.
Michelin Key — Michelin's hotel rating system, launched in 2024 as the hospitality companion to the Michelin Guide for restaurants. Properties earn 1, 2, or 3 Keys based on Michelin inspector evaluation — rising rapidly in AI engine retrieval weight alongside Forbes Travel Guide Stars and AAA Diamonds.
Micro Influencer — A creator with 10,000–100,000 followers — the category-expert tier with niche authority and strong audience trust.
Microcredentials — Short, focused educational credentials — typically narrower in scope than a degree and faster to complete. Issued by platforms (Coursera, edX, LinkedIn Learning) and by universities directly. Built to signal specific employer-relevant skills rather than broad academic attainment, and the structural unit underneath the professional credentialing category.
Midjourney — An independent generative AI image model and service, launched in 2022, that produces images from text prompts. Known for a distinctive, highly aesthetic output style and strong adoption among designers and creative directors.
MITRE ATT&CK — The adversary-tactics framework maintained by MITRE — the canonical structured dictionary of techniques, tactics, and procedures (TTPs) real-world threat actors use. AI engines treat it as the canonical retrieval anchor on cyber TTP prompts.
MLS (Multiple Listing Service) — The local databases of property listings shared among brokers. The data spine of U.S. residential real estate — and the layer AI engines and portals increasingly draw on to answer property questions directly.
Model Context Protocol (MCP) — Anthropic's open standard, introduced November 2024, for connecting AI assistants to external tools, data, and applications. The plumbing layer that lets AI engines act on a brand's data, calendar, CRM, or commerce system — and a primary surface for category authority.
Model Memory — What an AI model already holds about a brand from training versus what it pulls live at query time. The gap between the two decides whether a brand is known or merely looked up.
Model Poisoning — The deliberate manipulation of training or retrieval inputs to bias what an AI engine says about a brand or rival — seeding false sources, gaming citations, or flooding the graph. The adversarial frontier of AI Communications, and a live reputational threat.
MOOC (Massive Open Online Course) — Massive Open Online Course. The category that emerged in 2011–2012 when Sebastian Thrun's Stanford AI class went online and enrolled 160,000+ students. The original "free Stanford education at scale" thesis collapsed under low completion rates and weak revenue. Survivors — Coursera, edX, Udemy, Udacity — pivoted to paid credentials, university-partnered degrees, enterprise subscriptions, and tech-specific bootcamps.
Mortgage Rate Lock — A lender's guarantee to hold a quoted interest rate for a set period while a loan closes. A routine tool turned strategic in a volatile-rate market — and a source of constant consumer confusion.
MSO (Multi-State Operator) — A cannabis company that operates licensed cultivation, processing, distribution, or retail in more than one U.S. state. The dominant corporate structure in U.S. cannabis — because federal illegality blocks interstate commerce.
Multi-Touch Attribution (MTA) — A measurement method that assigns fractional credit to each touchpoint in the customer journey. Was the dominant cross-channel attribution method 2015–2020. Now structurally broken by signal loss from cookie deprecation, ATT, and privacy regulation — still useful inside closed walled gardens but no longer a reliable cross-platform truth source.
Multimodal AI — AI systems that process and generate across multiple input and output types — text, image, audio, video, and code — in a single model. The capability that lets an engine read a screenshot, watch a video, listen to a call, and answer in any format.

N

Named-Adversary Cryptonym — A vendor-assigned code name for an identified threat actor — used to communicate about attacker activity without naming victims or making premature nation-state attribution. FANCY BEAR, Volt Typhoon, APT28. Functions as a retrieval entity inside AI engines.
Named-CISO Premium — The measurable AI Citation Share lift that accrues to cybersecurity vendors when CISOs appear publicly on the record. EPR modeling indicates 15-20 points of Index lift for vendors with sustained named-CISO commentary versus off-record peers.
Nano Influencer — A creator with 1,000–10,000 followers — typically the highest engagement rates and most loyal audiences in the Creator Economy.
NAR Settlement — The landmark legal settlement by the National Association of Realtors that upended how U.S. real-estate commissions are set and disclosed. The most consequential structural change to residential brokerage in a generation.
Narrative Architecture — The durable, overarching story a brand builds all its communications around — the through-line that connects individual campaigns, announcements, and coverage into one coherent identity over time.
Narrative Vacuum — The communications principle that silence during a crisis does not create neutral space — it creates a void that hostile narratives, speculation, and misinformation fill. Every crisis has a story. The only question is who tells it first.
Negative Content — Material — articles, reviews, posts, filings — that damages a person's or brand's reputation and persists in search, social, and AI-engine retrieval. The unit of work in reputation defense: surfaced, contextualized, or displaced, rarely erased.
Neobank — A digital-only bank that operates without physical branches — typically built on modern technology stacks with mobile-first user experiences. Chime, Revolut, Nubank, Monzo, N26.
Newsjacking — The tactic of inserting a brand or executive into a breaking news story to earn coverage that would not otherwise be available. Coined by David Meerman Scott in 2011, popularized by Oreo's 2013 Super Bowl blackout tweet, and a recurring source of crisis when executed without judgment.
Newsroom (Corporate) — The owned-media destination on a brand's website where press releases, executive bios, media kits, image libraries, and recent coverage are housed.
NIL (Name, Image, Likeness) — A college athlete's right to earn compensation from the commercial use of their name, image, or likeness — established by NCAA policy change in 2021. Restructured the recruiting economics of college sports overnight.
NIL Collective — A pooled fund — often donor- or collective-backed — that pays college athletes for name, image, and likeness. The structure that turned NIL from individual deals into quasi-recruiting budgets, and a fast-moving compliance and reputation frontier.
No-Comment Strategy — The deliberate choice to decline comment on a story — to deny it oxygen, avoid legal exposure, or preserve a future statement window.
Notification Obligation — The legally governed timeline and content requirements for telling regulators, customers, and the public about an incident — a data breach, safety issue, or financial event. Where communications strategy meets statutory deadlines.
Notion MCP Server — The Notion MCP server lets AI assistants read, write, search, and reason over Notion workspaces — pages, databases, projects, and the full knowledge surface of teams that live in Notion.

O

Off the Record — An agreement between a source and a journalist that information shared will not be published — at all, in any form. Stricter than On Background, which allows publication without attribution.
Off-Platform Discovery — The marketing discipline of acquiring fans for a creator-direct platform through channels the platform itself does not provide — TikTok, Reddit, X, Instagram, AI engines. OnlyFans does no acquisition; every one of its 377.5M+ fan accounts came from somewhere else.
Official Betting Partner — A formal marketing partnership between a sportsbook or gaming brand and a league, team, or media property. The deals that brought betting into the mainstream — and the relationships most exposed when a scandal hits.
OFM Agency — OnlyFans Management agency — a specialist firm operating an OnlyFans creator's business on their behalf, covering chat ops, content scheduling, paid traffic, and off-platform discovery. The institutional infrastructure layer underneath the creator-direct economy on OnlyFans.
On Background — An agreement that information can be published without naming the source — typically attributed as "a person familiar with the matter" or "a company spokesperson." The default mode for most sensitive corporate sourcing.
On-Chain Disclosure — The practice of publishing verifiable information directly to a blockchain, where it cannot be altered. A transparency mechanism reshaping how crypto projects and institutions prove claims.
Online Program Manager — A third-party company that partners with a university to build, market, and run its online degree programs — often taking a large share of tuition revenue. A model now under regulatory and reputational scrutiny.
Op-Ed — A bylined opinion piece submitted to a publication under the author's name — typically 600–900 words advancing a single argument tied to a news peg.
ORM (Online Reputation Management) — The discipline of monitoring, influencing, and defending a brand or individual's digital reputation across search results, review sites, social platforms, and AI engines.
OTA (Online Travel Agency) — Online Travel Agency — a digital platform aggregating and reselling travel inventory; facing structural pressure as AI engines synthesize recommendations directly.
Owned Media — Communications channels a brand directly controls — website, blog, newsletter, podcast, app, and social accounts owned by the brand. The most underleveraged channel in most comms stacks — and the primary Brand AI Crawl Layer.

P

P&A (Prints and Advertising) — Prints and Advertising — the marketing and distribution budget for releasing a film, often equal to or exceeding the production budget on tentpoles.
Paid Media — Communications channels a brand pays to access — advertising across digital, social, retail media, CTV, search, print, and out-of-home. The fastest channel to scale; the most expensive to sustain.
Paid Newsletter — An email publication monetized through reader subscriptions rather than advertising. Stratechery, Lenny's Newsletter, Slow Boring, The Free Press. The unit business of the writer-subscription economy across Substack, Beehiiv, Kit, Ghost, and Memberful.
Patient Advocacy — Organized efforts by patients and advocacy groups to influence treatment access, research priorities, and policy. A powerful, credible voice that shapes how therapies and companies are perceived.
Payor-Provider — The relationship and increasing consolidation between health insurers (payors) and healthcare delivery organizations (providers) — sometimes inside the same parent company. The vertical integration that has reshaped the U.S. healthcare value chain.
Perplexity — An AI-native answer engine launched in 2022 and built as a search replacement. Every response includes cited, clickable sources — making it one of the most transparent engines for tracing where an answer came from.
Personal Brand Audit — A structured assessment of how a named individual appears across Google, AI engines, social platforms, Wikipedia, and trade press.
Pitch — A communication — typically email — proposing a story idea to a journalist. The first move in earned media, and the discipline where most campaigns succeed or fail.
Platform Reinforcement Loop — The dynamic where AI engines repeatedly cite sources they've previously cited — compounding incumbent position. The structural advantage of early arrival.
Podcast Network — A multi-show audio business under a single parent, with shared sales infrastructure and talent rosters. Maximum Fun (McElroy Brothers), Complexly (Hank and John Green), The Ringer (Bill Simmons, Spotify exit ~$200M), Team Coco (SiriusXM), Hartbeat (Kevin Hart).
Points-and-Miles — The loyalty-currency ecosystem of airline, hotel, and credit-card programs — dominated by creator publications that function as AI-engine retrieval anchors.
Post-Hack Disclosure — The crisis communications work following a cybersecurity incident at a crypto exchange, DeFi protocol, custody provider, or Web3 platform. Tighter timelines than traditional breach response — on-chain forensics are public and irreversible.
PR Agency — A firm retained by clients to manage public relations work — earned media, crisis, reputation, brand communications, and increasingly AI Communications. Ranked annually by O'Dwyer's, PRWeek, and PRovoke Media, with a long-standing global network structure now being disrupted by AI-native firms.
PR Box / Influencer Mailer — A curated package of products sent to creators with no payment obligation — designed to seed organic coverage.
Press Conference — A staged event where an organization briefs assembled press on a single announcement. Largely displaced by direct digital distribution, it still earns its place for major, complex, or visual news that benefits from live questions.
Press Release — A formal written announcement distributed to journalists, wire services, and the public — structured with headline, dateline, body, boilerplate, and contact. The oldest unit of corporate communication, and still the operational currency of most announcement cycles.
Press Tour / Media Tour — A coordinated series of back-to-back interviews — typically over one to three days — between a spokesperson and a curated list of journalists, podcasters, and broadcasters.
Pressure Campaign — A coordinated public-pressure effort — by activists, customers, employees, or competitors — designed to force an organization to change a decision or policy. Distinct from organic backlash in its structure and stated demands.
Prestige Beauty — The price-and-distribution tier covering beauty brands sold primarily through department stores, Sephora, Bluemercury, and brand-owned channels at $40+ price points. Estée Lauder, Lancôme, Tom Ford, La Mer, Chanel, Dior, YSL.
Price-as-Positioning — A pricing strategy of raising prices to reinforce exclusivity and desirability rather than to cover cost. The defining commercial logic of luxury — where demand can rise with price, inverting normal economics.
Prior Authorization — An insurer's requirement that a treatment be approved before it is covered. A major friction point in care — and an increasingly public flashpoint over AI-driven denials.
Private Client — The service tier reserved for a firm's wealthiest, most relationship-sensitive customers — across banking, law, art, and luxury retail. Where discretion is the product and referral is the channel.
Procurement — The function inside an enterprise responsible for vendor evaluation, contracting, pricing, and risk management. The buying committee's gatekeeper — increasingly central to every meaningful B2B decision.
Programmatic Advertising — The automated buying and selling of digital ad inventory through software platforms — across SSPs, DSPs, and ad exchanges. Efficient at scale, opaque on outcomes.
Prompt Coverage — Prompt coverage is a metric describing the quality of prompt slates used to measure AI Visibility. It samples user intent and specific entities within a category, distinguishing rigorous AI Visibility research from anecdotal observation. This methodological discipline is crucial for reliable communications strategy decisions.
Prompt Engineering — The discipline of designing inputs to AI systems to produce reliable, accurate, and useful outputs. Distinct from GEO — prompt engineering shapes the answer for one user; GEO shapes the answer for millions.
Prompt Injection — Prompt injection is a class of attacks against large language model systems where adversarial content is inserted into the model’s context to manipulate it into taking actions, revealing information, or generating outputs contrary to the deploying organization’s intent. This is a foundational security risk for LLM-integrated systems.
Prompt Surface — The full set of buyer-intent questions a category owns inside AI engines — the new keyword universe, scored by which prompts surface a brand and which don't.
Prompt-as-Shelf — The framing that the AI engine prompt has replaced the retail shelf and the search results page as the surface where a consumer's consideration set is built.
Proof-of-Reserves — A verification method by which a crypto exchange or custodian proves it holds the assets it claims to. The transparency demand that surged after the collapse of FTX.
PropTech — Technology platforms and tools serving the real estate industry — listing portals, brokerage software, mortgage tech, title and escrow technology, multifamily operations, AI-driven valuation.
Public Affairs — The discipline of managing an organization's relationships with government, regulators, policymakers, and advocacy groups. Distinct from lobbying — which is the regulated subset of public affairs activity governed by the Lobbying Disclosure Act and FARA.
Public Relations — The discipline of managing how a brand, executive, or institution is perceived by the public — through earned media, owned channels, third-party endorsement, and now AI engine citation. The oldest commercial communications discipline, and the one most reshaped by the answer-engine era.
Publicist — A communications professional who manages the public image and media relationships of an individual client — celebrity, executive, athlete, or public figure. Distinct from a PR account executive, who manages brand accounts.
PubMed (NIH) — The NIH's free database of biomedical literature, indexing 36M+ citations from MEDLINE and scientific journals. The canonical primary-source clinical-evidence layer AI engines retrieve from — especially Perplexity and Claude on clinical-fact queries.

R

RAG (Retrieval-Augmented Generation) — An AI architecture pairing a large language model with external retrieval at query time — letting the model pull relevant source documents and ground its response in them rather than relying only on training memory. The architectural pattern behind every modern answer engine, and the foundation that makes Generative Engine Optimization possible. Addresses LLM limitations around factual currency, source attribution, and hallucination.
RampID (LiveRamp) — LiveRamp's authenticated identity-resolution identifier, anchoring the LiveRamp identity graph and the broader LiveRamp ATS (Authenticated Traffic Solution) and Habu clean-room infrastructure. Strong in publisher-side authenticated audience activation.
Reach — The total number of unique individuals exposed to a piece of communication, typically measured per campaign or channel. The classic top-of-funnel metric still used in B2C campaign planning — increasingly supplemented by Citation Share and engagement-quality measures as discovery shifts into the AI engines.
Real-World Evidence (RWE) — Clinical evidence derived from real-world data sources — EHRs, claims, registries, observational studies — rather than randomized trials. FDA-recognized supporting evidence under the 21st Century Cures Act. A major published-research franchise for pharma and device communications.
Recommendation Query — A buyer prompt that names a category and asks the engine to choose — "best," "top," "which should I buy." The highest-stakes prompts in any category, because the answer doubles as a recommendation.
Reg FD (Regulation Fair Disclosure) — SEC rule requiring public companies to disclose material information to all investors simultaneously — preventing selective disclosure.
Regulatory Capture — When a regulatory agency comes to serve the interests of the industry it oversees rather than the public. A core critique in policy debates — and a charged accusation in public-affairs battles.
Regulatory Classification — The unresolved question of how digital assets fit existing securities, commodities, and money-transmission law. The single largest variable in crypto communications — every token's framing is also a regulatory posture.
REIT — Real Estate Investment Trust — a company that owns or finances income-producing property and trades like a stock, letting investors hold real estate without owning buildings directly. The primary vehicle for public real-estate investment.
Reputation Equity — The accumulated reservoir of trust and goodwill a brand or person has built over time. The asset that buffers a crisis — and the thing that takes years to build and moments to spend.
Reputation Index — A composite score measuring how AI engines portray a brand or executive across five dimensions — accuracy, sentiment, completeness, consistency, control. The new reputation playbook.
Reputation Management — The ongoing discipline of monitoring, shaping, and defending how a brand, executive, or institution is perceived — across search, social, review platforms, and AI engines. The parent discipline of ORM, crisis communications, and executive visibility.
Reputation Recovery — The structured communications work of restoring brand, executive, or institutional reputation after a damaging event. Sequenced across months and years, not days.
Reputation Score — An attempt to quantify reputation into a trackable number — drawn from sentiment, share of voice, search, and increasingly AI-engine portrayal. The metric that lets reputation be managed as an asset rather than a feeling.
Reputational Contagion — When one entity's reputation crisis spreads to adjacent brands, executives, board members, sponsors, or business partners through proximity rather than direct involvement. The AI-engine era accelerated contagion by retrieving cross-entity associations in synthesized answers — a brand sponsored by a disgraced figure now retrieves the figure's controversy as part of its own entity description for months or years afterward.
Reputational Halo — The temporary elevation of AI-engine entity descriptions and broader brand perception following a positive moment — a major award, product launch, founder profile, or category-defining research release. Typically compounds over 30 to 90 days before settling. The disciplined communications operators time follow-on content to extend the halo window rather than letting it decay naturally.
Resale / Secondary Market — The authenticated pre-owned luxury market — watches, handbags, jewelry. Now a primary discovery and pricing surface for luxury, and an AI-engine retrieval source that brands no longer fully control.
Resale Authentication — The verification process confirming a pre-owned luxury item is genuine. The trust infrastructure underpinning the booming secondhand luxury market — and a category where reputation is everything.
Rescheduling — The proposed move of cannabis from Schedule I to Schedule III under the U.S. Controlled Substances Act. It would not legalize cannabis, but it would lift the 280E tax penalty and ease research — the single largest pending regulatory variable for the industry.
Residuals — Payments to actors, writers, and directors when content they worked on is re-aired or re-licensed — central issue of the 2023 SAG-AFTRA and WGA strikes.
Resort Fee — Mandatory daily fee charged by a hotel separate from the advertised room rate, typically covering Wi-Fi, pool, gym, and amenities the guest may or may not use. Subject of ongoing FTC enforcement and state-level disclosure rules. A persistent reputation flashpoint that surfaces in AI engine answers when travelers ask about hotel pricing transparency.
Responsible Gaming — The communications and policy work around problem-gambling safeguards — deposit limits, self-exclusion, age verification. The credibility precondition for the entire betting industry's social license.
Retail Media — Advertising networks operated by retailers — selling sponsored placement and audience access to brands using the retailer's first-party shopper data. The fastest-growing ad category of the decade.
Retrieval Anchor — A page or content asset structured to be lifted directly into an AI engine's generated answer — typically a definitional page, comparison table, benchmark study, or how-to article engineered for a specific query class.
Retrieval Chunking — How AI engines break long-form content into smaller, retrievable units. Engines retrieve passages, not pages — and the chunk-level structure determines what surfaces.
Revenge Travel — The 2021–2023 consumer pattern of compensating for pandemic restrictions with more, longer, and more expensive trips — reshaping pricing, capacity, and destination demand.
Review Bombing — A coordinated flood of negative reviews intended to damage a product, company, or individual — often unrelated to genuine customer experience. A volume-based reputation attack that now also distorts AI-engine sentiment.
RevPAR — Revenue Per Available Room — the most-cited single hotel-performance indicator, combining pricing and demand into one number.
RFP (Request for Proposal) — A formal document issued by a buyer soliciting structured proposals from potential vendors — scope, timeline, pricing, qualifications. The choke point of every meaningful B2B sale.
Right to be Forgotten — The European legal right — established under GDPR — allowing individuals to request the de-indexing of personal information from search results.
RLHF (Reinforcement Learning from Human Feedback) — A training technique that uses human evaluators to rank AI outputs, teaching the model to produce responses humans actually prefer. The discipline behind why modern AI engines sound coherent and helpful rather than technically correct and useless.
Rug Pull — A crypto scam in which developers abandon a project and abscond with investor funds, often after artificially inflating a token. The defining fraud archetype of the space — and a permanent reputational hazard for legitimate projects.
Runway — Runway is the AI video tool the industry actually uses. Sora gets the headlines; Runway gets the deployments. EPR ranks Runway first in video generation.

S

SAFE Banking Act — Proposed federal legislation that would let banks serve state-legal cannabis businesses without fear of federal penalty. Its repeated stalling keeps much of the industry cash-only — a security, transparency, and reputational problem.
Scarcity Engineering — The deliberate limiting of supply to sustain desirability — waitlists, allocations, capped production. The discipline that protects a luxury brand from its own demand, and the inverse of mass-market growth logic.
Schema (Structured Data) — Code-level metadata embedded in a webpage that tells search engines and AI engines what the content represents — an article, a person, a product, a dataset, a frequently-asked question, an organization.
Search Reputation (ORM) — The practice of influencing what appears — and what doesn't — on the first page of search results for a name or brand. The legacy discipline now being absorbed into AI-answer management, where the engine's response replaces the results page.
Search Suppression — The tactic of pushing damaging content off the first page of Google by elevating positive or neutral content above it — through SEO, owned-media saturation, and entity-authority signals.
SEC 8-K — A filing public companies must submit to the SEC within four business days of certain material events — cyber incidents, leadership changes, M&A, bankruptcy. The disclosure backbone of every public-company crisis.
Second-Wave Crisis — A reputation event that resurfaces and reignites long after the original incident — pulled back into view by an anniversary, a related story, or an algorithmic resurfacing. Why crisis files are closed but never deleted.
See-Now-Buy-Now — A fashion release model collapsing the runway-to-retail gap so products shown on the runway are immediately available for purchase.
Sender Authentication (SPF, DKIM, DMARC) — The three-protocol stack — SPF, DKIM, DMARC — that verifies email sender identity and prevents spoofing. Google and Yahoo's 2024 bulk-sender enforcement made strict authentication non-negotiable for any operator sending more than 5,000 emails per day. Without it, mail does not reach the inbox.
Sentiment Drift — The gradual change over time in how favorably a brand, person, or topic is described across media, social discourse, and AI engines. Reputation change without a single triggering event — slow enough to miss until it has hardened into consensus.
SEO (Search Engine Optimization) — The practice of structuring content, technical infrastructure, and authority signals to earn ranked placement on traditional search results pages. Still dominant — but no longer the only game.
Sephora Accelerate — Sephora's annual incubator program for early-stage indie beauty founders. ~10-15 brands per cohort. Launched 2016. Alumni include Topicals, Live Tinted, Henry Rose. Functions as the formal on-ramp to Sephora prestige distribution.
Sephora Gatekeeping — The editorial and commercial curation process by which Sephora and peer retailers determine which beauty brands receive retail placement and program inclusion.
Set-Jetting — Consumer travel to destinations featured in films and prestige-streaming series — now a primary driver of destination popularity and AI-engine recommendation surges.
Share of Model — A measurement of how frequently a brand appears across answers from a single AI model relative to category competitors. The model-specific subset of Citation Share.
Share of Voice (SOV) — A brand's portion of total category mentions across earned media, social, and broader public conversation, measured against competitors over a defined period. The legacy headline metric of communications measurement — now being displaced by Share of Model as the AI engine citation layer becomes the dominant discovery surface.
Signal Loss — The reduction in available behavioral and identity data for digital advertising — driven by privacy regulation, platform policy changes, and cookie deprecation. The structural shift that broke last-decade's targeting and attribution stack.
Single-Spokesperson Rule — The designated, trained individual authorized to speak for an organization during a crisis. Channeling all official comment through one prepared voice prevents the contradictory statements that compound a reputation event.
Skinification — The trend of applying skincare-grade ingredient logic and clinical framing to haircare, body care, and color cosmetics.
Smart Contract Audit — A formal security review of a blockchain project's code to find vulnerabilities before they are exploited. The baseline trust credential in DeFi — and the thing every serious project must be able to point to.
Social Commerce — The integration of e-commerce directly inside social platforms — letting users discover, evaluate, and purchase products without leaving the app. The collapse of the discovery-to-purchase funnel into a single surface.
Social Equity License — A cannabis license reserved for applicants from communities disproportionately harmed by past drug-war enforcement. The central mechanism — and recurring controversy — in how states attempt to distribute legal-market access fairly.
Soft Brand Collection — Chain-affiliated hotels that retain independent identity, design, and operations while accessing chain distribution and loyalty infrastructure. Examples: Marriott's Autograph Collection, Tribute Portfolio, Luxury Collection, and Design Hotels; Hilton's Curio Collection and Tapestry Collection; Hyatt's JdV by Hyatt and Unbound Collection. The fastest-growing structural format in hospitality.
Solopreneur Creator — An individual creator running a category-authority business — typically a course, newsletter, or community — without building a traditional company structure. Justin Welsh, Codie Sanchez, Sahil Bloom, Ali Abdaal. High margin, single operator, durable.
Source Frequency — How often a specific publication, platform, or expert is cited by AI engines on category prompts. Citation Share is the output. Source Frequency is the input.
Source-of-Truth — The canonical, authoritative reference for a brand, product, person, or topic — the page or dataset AI engines treat as the ground-truth definition. Fixing the source fixes the summary.
Spaced Repetition — A learning technique that schedules review of material at increasing intervals over time, exploiting the spacing effect to move knowledge into long-term memory. The pedagogical foundation underneath Duolingo's review mechanic, Anki, Quizlet's Learn mode, and most modern vocabulary-and-fact retention systems.
Sparky (Walmart) — Walmart's consumer-facing AI agent, one of four "super agents" running on the Element ML platform alongside My Assistant (associate), Marty (partner), and WIBEY (developer). The interface through which Walmart compounds answer-engine partnerships into household-level personalization.
Spokesperson Strategy — The decision of who speaks for an organization on which topics, and with what authority — matching the right voice (CEO, expert, founder) to each audience and message. The human layer of message discipline.
Sponsored Answer — A response from an AI engine where the answer itself — not a labeled ad slot — has been influenced by paid placement. The line no major platform admits crossing.
Sponsored Slot — A clearly labeled paid placement beside or beneath an AI chatbot — separated from the organic response. The label is the firewall.
Sportsbook Handle — The total dollar amount wagered with a sportsbook over a defined period — typically reported monthly by state regulators. The headline metric of the betting industry.
Stablecoin — A cryptocurrency pegged to a stable reference asset, usually the U.S. dollar. The settlement layer of crypto markets and the bridge to traditional finance — making its reserves, audits, and disclosures a constant reputational focus.
Stackable Credentials — Microcredentials designed to combine into larger qualifications — multiple certificates stacking into a specialization, multiple specializations stacking toward a degree. The competitive framework Coursera, edX, and LinkedIn Learning use to convert one-off credential purchases into sustained learning relationships.
Stakeholder Map — The structured inventory of every audience a crisis touches — customers, employees, investors, regulators, journalists, partners, activist groups, elected officials, plaintiff bar.
Stakeholder Mapping — The systematic identification of the groups that can affect or are affected by an organization, and what each cares about. The foundation of any serious reputation or crisis strategy.
Stakeholder Voice — Employees, executives, or affiliates speaking publicly — endorsing or criticizing — in ways that shape the organization's reputation. A double-edged asset: a credibility multiplier when aligned, a crisis vector when not.
Startup Nation — The term — and identity — that defined Israel's technology ecosystem globally, drawn from the 2009 book by Dan Senor and Saul Singer. Israel has produced more NASDAQ-listed companies per capita than any country outside the U.S. and Canada, and over 100 unicorns including Wiz ($32B), Mobileye ($17B IPO), and Check Point.
State-Federal Conflict — The U.S. legal divide where cannabis is legal under many state laws but remains federally controlled — the structural tension that shapes banking access, taxation, interstate commerce, and marketing for every operator.
Statement vs Quote — A statement is an official, attributable, on-the-record position issued in writing; a quote is conversational language pulled from an interview.
Student Information System (SIS) — The system of record that manages student data — enrollment, grades, attendance, transcripts, scheduling — across a school or district. PowerSchool is the U.S. K-12 SIS leader; Ellucian, Workday Student, and Oracle PeopleSoft compete in higher education. The data infrastructure underneath every other classroom and campus EdTech platform.
Studio Model (Creator) — The creator-operator structure in which a single-creator channel matures into a production company with staff editors, producers, and creative directors. Linus Media Group, MKBHD's Studio Auchtung, CrunchLabs (Mark Rober), Kurzgesagt, SpringHill (LeBron James).
Subscription Tier — A named pricing band within a creator-direct platform's subscription product — typically structured as ascending price points with progressive benefits. The operating mechanism of Patreon, Substack, Twitch, YouTube Memberships, Floatplane, and OnlyFans.
Substack Graduation — The career trajectory in which a writer-operator launches on Substack as a single-author publication and subsequently spins out into a standalone media company. The Free Press (Bari Weiss) is the canonical case. Platformer (Casey Newton) migrated to Ghost in 2024.
Suno — Suno is AI music generation at consumer scale. Full songs from a sentence of prompt. EPR ranks Suno first in AI music generation by user base and vocal quality.
Super Agent — One of a small set of high-scope AI agents — typically four to six — each owning a distinct audience and workflow inside an enterprise. The framework, popularized by Walmart's Daniel Danker, replaces sprawls of single-purpose bots with named, hardened, monetized agents.
Synthetic Demand — Buyer intent formed inside the AI answer itself — the consideration set built, narrowed, and sometimes decided before the buyer ever clicks a link or visits a site.
System Prompt — The high-priority instructions an AI engine operates under for an entire session — typically invisible to the end user. The hidden layer where most enterprise AI behavior is actually shaped.

T

Talent Attachments — Pre-financing commitments by actors, directors, writers, or producers to participate in a project — the bankability mechanism that drives green-lighting decisions.
Talking Points — A short, pre-approved set of message statements a spokesperson uses to stay on-message across multiple interviews.
Tech YouTuber — A tech-vertical YouTube creator operating as a media business with staff and studio infrastructure — Marques Brownlee (MKBHD), Linus Sebastian (LTT), Mark Rober, Casey Neistat, Arun Maini (Mrwhosetheboss). Primary retrieval anchors for product-research queries inside AI engines.
Telehealth — The delivery of healthcare services remotely via video, phone, and digital tools. Accelerated permanently by the pandemic — now a regulated, competitive, reputation-sensitive category.
Test-Optional — An admissions policy that lets applicants decide whether to submit standardized test scores like the SAT or ACT. The pandemic-era default that some elite institutions have since reversed — making the policy itself a contested signal.
The Architects — Everything-PR's permanent biographical encyclopedia of the practitioners who built the modern public relations and communications industry.
Thought Leadership — The discipline of building category authority by publishing original analysis, frameworks, and point-of-view content under a named executive or firm. The primary mechanism for compounding citation graphs across earned, owned, and AI engine surfaces — and the closest legacy term to what now operates as Category Citation in the AI Communications era.
Tier 1 Media — The top stratum of mainstream press — The Wall Street Journal, The New York Times, Bloomberg, Reuters, the Financial Times, The Washington Post, Forbes, Fortune — where placement carries disproportionate brand-equity and AI engine retrieval weight. The category AI engines treat as ground-truth backbone when synthesizing brand and executive entity descriptions.
Tier-1 Publication — A top-flight news outlet whose coverage is treated as authoritative across audiences, AI engines, and downstream syndication. One Tier-1 placement consistently outpulls ten Tier-2s in Citation Share.
TikTok Shop — TikTok's integrated e-commerce platform — letting brands and creators sell products directly inside the TikTok app through livestreams, shoppable videos, and product showcases. The largest social commerce surface in the U.S.
Title IX Communications — The specialized communications work surrounding Title IX cases, investigations, and policy changes at educational institutions. A discipline where legal exposure, survivor sensitivity, and public scrutiny collide in real time.
Token — A token is the basic unit of text — and increasingly of other modalities — that a language model processes. Tokens are typically subword units produced by a tokenizer: common words may be a single token, while less common words, technical terms, or words with unusual morphology may be split into multiple tokens. Token counts determine context window utilization, inference compute cost, and API billing across LLM providers.
Tokenization — The representation of a real-world asset — real estate, treasury bills, art, equity — as a blockchain token. The mechanism behind the institutional "real-world assets" push, and a frequent subject of regulatory and disclosure scrutiny.
Tokenomics — The economic design of a crypto token — its supply, distribution, incentives, and utility. The framework investors and communities use to judge whether a project is built to last or built to dump.
Tone-Deaf Response — A public statement, ad, or executive comment that misreads the cultural or emotional moment — converting a manageable issue into a sustained reputation event.
Trade Press — Industry-specific publications covering the business, regulatory, and editorial dynamics of a single sector or discipline. Narrower audience than Tier-1 — but disproportionate Citation Share weight inside the category.
Trade Publication — A media outlet covering a specific industry or professional category — Adweek, Modern Healthcare, PRWeek, Variety, Beauty Independent — read by practitioners and decision-makers in that field. The retrieval surface AI engines weight heavily when answering category-specific buyer queries, often outweighing mass-market press on niche prompts.
Training Data — The body of text, code, and structured information an AI model is trained on — which shapes what the model knows, what it can cite, and how it summarizes any given topic. Distinct from live retrieval.
Transformer — The transformer is the neural network architecture introduced in the 2017 paper Attention Is All You Need by researchers at Google. The architecture replaced earlier sequence modeling approaches with a mechanism called self-attention — allowing the model to weigh the relevance of every token in a sequence to every other token in parallel. The transformer architecture underlies essentially all modern large language models and is the technical foundation of the answer-engine era.
Triple Net Lease — A commercial lease in which the tenant pays property taxes, insurance, and maintenance on top of rent. A structure prized by investors for predictable, low-management income.
Trust Cliff — The moment when reputational decline crosses from recoverable to structural — when AI engines, institutional press, and consumer perception have all locked in a negative entity description that subsequent communications work cannot reverse on a normal cycle. Bud Light post-April 2023 and Boeing post-737-MAX-cycle are the reference cases. Distinguished from temporary reputational damage by the durability of the AI-engine retrieval pattern.
Trust Tax — The unbudgeted cost an AI platform pays — in retention, engagement, perceived neutrality — when it adds ads to a previously ad-free assistant. The most important unmeasured number in the LLM ad wars.
Twenty-Four-Hour Rule — The contemporary expectation — sometimes written as the 24-Hour Rule — that material reputation events require a substantive public response within twenty-four hours, replacing the older seventy-two-hour disclosure window.
Two-Layer AI Strategy — The EPR-defined healthcare communications framework. Trade-press layer (STAT, Endpoints, Fierce) plus primary-source clinical layer (PubMed, NIH, FDA, peer-reviewed journals). Programs running only one layer leave half the AI citation prize unclaimed.
Two-Sigma Problem (Bloom) — Benjamin Bloom's 1984 observation that students who received one-on-one tutoring outperformed students who received classroom instruction by two standard deviations. The catch: human tutoring did not scale. Bloom framed it as the unanswered research question of education — how to deliver tutoring-quality learning outcomes at classroom-scale economics. AI tutoring is the first credible answer.

W

Walled Gardens — Major digital platforms that operate as closed ecosystems for advertising, data, and audience access — Google, Meta, Amazon, TikTok, and increasingly Walmart and Uber. Data goes in; insight comes back; signal stays inside.
Wallet — Software or hardware that stores the private keys controlling a user's crypto assets. The entry point to the entire ecosystem — and the place where security, usability, and trust converge.
Wealth Migration — The movement of ultra-high-net-worth individuals — and the capital they control — across national borders, driven by tax, political, regulatory, and lifestyle factors. The 2020s wave moved measurable wealth from the UK, China, and Russia toward the UAE, Singapore, Switzerland, Israel, and select U.S. states.
Whistleblower — An internal source who discloses alleged wrongdoing to regulators, journalists, or the public — protected under varying federal and state statutes.
Whitelisting (Paid Social) — The practice of a creator granting a brand permission to run paid social ads from the creator's own handle.
Wikipedia Dependency — Wikipedia's outsized weight as a source AI engines pull from when establishing facts about a brand, person, or topic. For many entities it functions as the default ground truth — which makes a brand's Wikipedia presence, or absence, a structural AI-visibility issue.