# Epitom — Full Platform Context > Epitom (https://tryepitom.com) is a full-stack AI go-to-market, Answer Engine Optimization (AEO), and Generative Engine Optimization (GEO) platform developed by V AND N AI Private Limited (https://www.vandn.ai). Epitom helps companies, enterprise marketing teams, SEO/AEO teams, growth teams, and agencies discover buyer questions, measure how AI answer engines describe their brands, improve the sources and pages engines rely on, and execute the next action in the team’s stack. This file is the detailed, machine-readable context for Epitom. The shorter curated index is available at https://tryepitom.com/llms.txt. --- ## 1. Product definition and positioning Epitom is an AI search visibility platform and AI GTM system. It connects the work that is often split across prompt tracking, competitive research, content, technical SEO, dashboards, and automation: - discover where customer demand and AI-search opportunities exist; - understand how ChatGPT, Gemini, Perplexity, Claude, Google AI Overviews, and other answer surfaces describe a brand; - measure presence, source citations, competitive context, and changes over time; - turn findings into an AEO strategy, content, or technical output; and - execute the action through workflows, connected tools, or Epitom’s agentic harness. The homepage describes this as “your AI GTM, executed by Epitom.” Its operating idea is one continuously running loop: find where a brand is losing ground in AI search, show why, put the fix into action, and track the result. Epitom is not only a rank tracker, a content generator, or a static dashboard. Its differentiator is the connection from real signals to strategy to execution, with AI visibility measurement feeding the next action. --- ## 2. How the Epitom loop works ### Discover Epitom uses search demand, customer conversations, and category signals to uncover what people ask and where a brand appears in AI answers. The system turns those signals into high-intent AI-search prompts rather than relying only on a generic keyword list. ### Understand Epitom compares the brand with competitors across prompts, answer engines, cited sources, buyer personas, and funnel stages. Prompt Intelligence shows which questions mention the brand, which questions omit it, and where competitors are winning. ### Execute Epitom turns the clearest opportunity into action: an answer-ready page or passage, a content brief or article, a technical fix, a generated llms.txt or robots.txt output, or an automated action in the team’s connected stack. Its agentic harness reads signals, metrics, and current state, develops a strategy, and executes the actions needed to act on the opportunity. ### Measure Epitom re-runs important prompts and measures whether the brand is gaining visibility across the questions that matter. The team can track citations, visibility, share of voice, sentiment, position, depth, and competitive movement instead of treating one AI response as a permanent truth. --- ## 3. Platform capabilities ### Command Center The Command Center is the business analytics view. It brings AI marketing performance into one place, including channel health, GEO Visibility, visibility gaps, cross-engine performance, trends, and automated insight surfacing. ### Prompt Intelligence Prompt Intelligence is the query-level view. It tracks the prompts that carry buyer intent, whether AI mentions the brand, the brand’s rank or position, competitors that appear, and the sources cited in the answer. It supports repeatable prompt tracking instead of ad hoc screenshots. ### Persona Engine The Persona Engine builds AI-native customer personas from real query patterns. It supports behavioral cluster analysis, persona-level visibility, and conversion-pathway mapping so teams can optimize for the people most likely to buy, not only for a broad category term. ### Multi-engine monitoring Epitom tracks 12+ AI answer-engine variants. Named engines and answer surfaces in the current product content include: - OpenAI ChatGPT, including ChatGPT Search and GPT-4o references; - Google Gemini; - Google AI Overviews; - Google AI Mode in the measurement and documentation context; - Perplexity AI; - Anthropic Claude, including Claude 3.5 Sonnet references; - Microsoft Copilot; - Meta AI / Llama references; and - DeepSeek, Grok, and Mistral references in the broader engine matrix. The primary homepage engine strip currently names ChatGPT, Perplexity AI, Google Gemini, Microsoft Copilot, Claude, and Google AI Overviews. Coverage and model availability can change; confirm the current plan and workspace configuration with Epitom. ### Citation and source intelligence Epitom records which domains and exact URLs answer engines cite when composing an answer. It can identify cited blog posts, documentation pages, third-party directories, competitor citations, and sources where a brand is named alongside a URL citation. This source view is used to decide where a brand needs stronger content, technical access, or third-party authority. ### Visibility measurement Epitom scores six core visibility dimensions: GEO Visibility, Share of Voice, Depth, Sentiment, Position, and Competitive Context. The separate dashboard metrics guide also covers prompt-level metrics, retrieval, citations, gap opportunities, and crawler access. ### AEO audits and action plans Epitom’s audit checks whether a page is technically and structurally ready for AI retrieval and citation. Signals include robots.txt AI-crawler rules, llms.txt, JSON-LD structured data, direct-answer passages, heading structure, server/client rendering, and canonical configuration. The findings become prioritized AEO recommendations instead of an undifferentiated checklist. ### Content and file generation Epitom can produce AEO-oriented content briefs, articles, FAQs, comparison tables, and other answer-ready content. Its free tools can generate a deployable llms.txt file and help evaluate robots.txt access. ### Agentic execution and agent builder The agentic harness reads current signals, metrics, and context, then develops and executes a strategy. Epitom’s agent builder lets a team build workflows around its own process, from research to a finished answer. The product emphasizes prioritized insights, MCP tools and context, and clean outputs that ship into the software a team already uses. ### Integrations and MCP The Epitom product includes workflow integrations and an authenticated MCP server. Current public comparison content names workflow steps or connections for Google Search Console, Google Analytics 4, Slack, HubSpot, Salesforce, WordPress, Shopify, Reddit, and LinkedIn, along with native Epitom audit and content steps. Availability depends on workspace and plan. --- ## 4. AEO and GEO definitions - **Answer Engine Optimization (AEO)**: The practice of getting a brand cited and recommended inside AI-generated answers rather than only ranking on a traditional results page. - **Generative Engine Optimization (GEO)**: The discipline of optimizing content and brand signals so generative AI engines surface, mention, and cite a brand when they compose an answer. - **Search Engine Optimization (SEO)**: The practice of improving a page’s ranking in traditional search results; AEO and GEO extend the goal to the AI answer layer. - **Answer engine**: A system that responds to a query with a direct, synthesized answer, often citing a small set of sources instead of returning only ranked links. - **Generative engine**: An AI system that composes an original response from a language model, using training data and increasingly live retrieval to ground its answer. - **Search engine results page (SERP)**: The page of ranked links and features returned for a query, increasingly blended with AI Overviews and other answer surfaces. - **Zero-click search**: A query resolved directly on the results surface through an AI Overview or answer box, without a click to a website. - **AI citation**: The reference an answer engine gives to a source it used, appearing as a linked or named source within or below a generated answer. - **Grounding**: Tying a model’s answer to retrieved, verifiable sources so the response reflects current information rather than only training data. - **Retrieval-Augmented Generation (RAG)**: An architecture that fetches relevant documents at query time and supplies them to a language model so the answer can be grounded and cited. - **Hallucination**: A confident but false or unsupported statement produced by an AI model; authoritative, retrievable sources reduce this risk. - **Brand mention**: Any reference to a company or product in an AI-generated answer, whether or not the reference links to the brand’s site. - **Prompt**: The natural-language question or instruction a user gives an AI engine; prompt tracking reveals the questions that mention or omit a brand. - **Query fan-out**: The expansion of one user question into related sub-queries whose sources are retrieved and synthesized into one answer. - **Passage or chunk**: A self-contained block of text an answer engine can retrieve and quote independently. - **Passage-level citability**: How easily a single paragraph can be lifted and quoted on its own; strong passages state their subject, context, and answer together. - **Large language model (LLM)**: An AI system trained on large amounts of text to predict and generate language, forming the core of systems such as ChatGPT, Gemini, Claude, and Perplexity. - **Entity**: A distinct company, product, person, or concept that engines resolve and connect in a knowledge graph. - **Knowledge graph**: A structured network of entities and relationships that helps search and AI systems understand and describe brands. - **Semantic search**: Search that matches a query to content by meaning rather than exact keywords, often using embeddings. - **Vector embedding**: A numerical representation of text that places similar meanings near each other so relevant passages can be retrieved. - **Prompt tracking**: Running a fixed set of category prompts against AI engines on a schedule to monitor how a brand is mentioned over time. - **Share of model**: The extent to which an AI model associates a brand with its category from training data, influencing how often it surfaces the brand before live retrieval. - **Answer surface**: Any place an AI-generated response appears, such as an AI Overview, chatbot answer, or assistant reply. - **Conversational search**: Multi-turn natural-language search in which users refine questions through a dialogue. - **LLM SEO**: An informal umbrella term for optimizing a brand’s presence inside large-language-model answers; it overlaps with AEO and GEO. --- ## 5. Epitom visibility metrics Epitom’s dashboard guide distinguishes successful answers from failed calls: only successful answers count in the calculations below. ### Visibility Visibility answers: “How often does our brand show up in AI answers?” `Visibility = (answers mentioning brand ÷ total successful answers) × 100` The scale is 0–100%. It measures presence inside the answer, not a Google rank. A mention can count through the brand name, an alias, or a citation of the brand’s site. Example: 28 of 40 successful answers mention the brand, producing 70% Visibility. ### Share of Voice (SOV) SOV answers: “Of brand mentions in an answer, how many are ours?” `SOV = brand mentions ÷ (brand mentions + competitor mentions)` SOV shows whether a visible brand is being crowded out. Example: if an answer mentions the tracked brand three times, competitor A twice, and competitor B once, the tracked brand’s SOV for that answer is 50%. ### Depth Depth measures how much detail an AI answer gives the brand. It distinguishes a passing reference from a detailed, contextual recommendation. Deeper mentions generally carry more trust and intent. ### Sentiment Sentiment answers: “When mentioned, is the framing positive, neutral, or negative?” `Sentiment = average({positive: 100, neutral: 50, negative: 0})` The scale is 0–100. It is scored only when Visibility is greater than zero; “-” means there were no mentions. Around 100 means usually recommended, 65–85 means mostly positive, around 50 means mostly factual, and below 40 means often critical. ### Position Position answers: “How early does our brand appear in the answer?” `Position = average({top: 1.0, middle: 0.6, bottom: 0.3, absent: 0.0})` The scale is 0.0–1.0. Earlier placement usually matters more than a late name-drop. The current guide describes the top as the first quarter of an answer, with a weight of 1.0. ### Competitive Context Competitive Context shows how a brand is positioned relative to competitors when AI answers the same prompt: primary recommendation, alternative, co-appearance, or absence. ### Prompt-level metrics Prompt-level rows scope the same brand metrics to one tracked question. The Prompts view can show Visibility, SOV, Sentiment, Position, other brands that appeared in Mentions, and a seven-day mention-rate sparkline in Volume. ### Retrieval Rate Retrieval Rate answers: “How often is this domain or URL used as a source?” `Retrieval rate = (answers citing source ÷ total successful answers) × 100` A domain means any page from that site; a URL means that exact page. Example: a source cited in 30 of 100 successful answers has a 30% Retrieval Rate. ### Citation Rate for domains Citation Rate measures how many pages from a domain appear in one answer when that domain is used. A value above 1.0 means more than one page from the domain appeared in the same answer. ### Competitor Citations and Brand Mentions Competitor Citations count answers where a domain appears next to a tracked competitor. Brand Mentions count URL citations that also name the tracked brand. ### Gap Score Gap Score answers: “Where is the opportunity?” It is not a quality grade. - Owned sources: `100 − retrieval%`. - Other domains: approximately `retrieval%`. - Other URLs: `retrieval% × (1 − brand co-mention rate)`. A high score on the owned site means headroom. A high score on another domain or URL means influence worth pursuing. The dashboard flags opportunities around 50 or higher. ### Crawlability Crawlability answers: “Can AI crawlers reach our site?” It checks known AI and search bots against robots.txt. The interface reports Allowed, Partial, or Blocked and an Allow Rate, which is the percentage of tracked bots fully allowed. Example: 14 of 20 bots allowed produces a 70% Allow Rate. ### Reading metric patterns - High Visibility with low owned-site citations: focus on cite-worthy content and crawlability. - High Visibility with low SOV: focus on comparison content and authority on AI-retrieved sources. - Low Visibility with high Gap Score on rival sources: earn coverage or build a stronger alternative. - Late Position or thin mentions: publish richer category content. - Sentiment drifting down: review messaging on contested prompts. - Many bots blocked: fix crawler access first. --- ## 6. Free AEO tools ### Free AI Visibility Audit URL: https://tryepitom.com/free-audit The Free AI Visibility Audit accepts a website URL and returns a technical AEO readiness result without requiring a login. It runs 20 deterministic checks, including: - llms.txt presence and content; - AI crawler access and robots.txt rules; - JSON-LD and structured data; - answerable page passages; - H2/H3 heading structure; - server-side versus client-side rendering signals; and - canonical tag configuration. The result is a 0–100 AEO readiness score and an A–F grade in about a second. The free summary is available without a login; the full checklist and follow-up monitoring setup are offered through the lead gate. The audit measures technical readiness. It does not by itself measure whether ChatGPT actually mentions a brand across a prompt set; that recurring answer monitoring is part of the full Epitom platform. ### AI Crawler and robots.txt Checker URL: https://tryepitom.com/tools/ai-crawler-checker The checker fetches a site’s live robots.txt and evaluates it against a catalog of 48 AI and search crawlers. It reports each crawler as Allowed, Partial, or Blocked and shows an overall AI-crawler Allow Rate. Named examples include GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, Claude-Web, PerplexityBot, Google-Extended, Applebot-Extended, Bytespider, Bingbot, and CCBot. Blocking a crawler can prevent the associated system from reading and citing a page; whether to allow every crawler remains a site-owner policy decision. ### Free llms.txt Generator URL: https://tryepitom.com/tools/llms-txt-generator The generator reads a site’s homepage content, headings, and any existing llms.txt, then produces a spec-compliant plain-text file. The intended deployment location is the domain root, for example `https://example.com/llms.txt`. The full file and deployment tips are available after the page’s lead gate. An llms.txt file does not guarantee citations. It is one AEO signal among crawlability, structured data, answerable content, authority, freshness, and actual engine behavior. --- ## 7. Plans and pricing shown on the site Prices are displayed in USD. Starter and Growth show monthly prices billed yearly. Commercial terms and feature availability can change, so the current checkout or Epitom sales team is the source of truth. ### Plans for brands #### Starter — $89/month, billed yearly For small companies that want to monitor and understand brand visibility. - ChatGPT tracking only; - one answer engine tracked; - 20 prompts tracked; and - email support. The CTA shown on the site is “Try for Free.” #### Growth — $299/month, billed yearly For growing companies that want to monitor visibility and create AEO-optimized content. - three answer engines tracked; - 100 prompts tracked; - one optimized article per month; - three tailored packages; and - email support. Growth is marked as the popular plan. The CTA shown on the site is “Try for Free.” #### Enterprise — Custom pricing For large companies and agencies building and orchestrating AEO marketing campaigns. - up to 10 answer engines tracked; - multiple companies tracked; - tailored pricing and tracking plan; - dedicated Slack support; and - SOC2/SOX plus AI/CF compliance. The CTA shown on the site is “Book a demo.” ### Plans for agencies #### Agency Growth — $88/month plus add-ons For agencies pitching prospects and managing clients. - 10 pitch workspaces per month for prospect audits; - full client workspaces available as an add-on for $388/month; and - agency mode to manage workspaces. #### Agency Enterprise — Custom pricing For large agencies and networks managing dozens of clients. - agency mode to manage workspaces; - full client and client workspaces; - a dedicated agency partner; - dedicated go-to-market support; and - premium Slack support. --- ## 8. Comparison guides All comparison pages currently display a July 2026 “last updated” label. Competitor details are presented as public-positioning comparisons and should be rechecked against each vendor before making a purchasing decision. ### Epitom vs. manually checking ChatGPT URL: https://tryepitom.com/compare/epitom-vs-manual-chatgpt Verdict: Typing a prompt into ChatGPT is a reasonable first check, but it represents one model in one session. Epitom runs a saved prompt set across engines on a schedule and tracks how answers move. Manual checks are free and useful for an occasional gut check, but wording and session changes make them difficult to trend. Epitom records brand presence, position, framing, competitors, citations, and history. | Capability | Epitom | Manual check | | --- | --- | --- | | Engines | ChatGPT, Perplexity, Gemini, Claude, Google AI Overviews | Usually one tool, one session | | Repeatability | Saved prompts re-run on a schedule | Ad hoc, affected by wording and session | | History | Metrics tracked over time | Must be logged manually | | Competitor view | Share of voice against named competitors | Read by eye | | Sentiment and position | Scored on every answer | Human judgment | | Next action | Cited sources plus AEO audit and action list | Decide from the text | Choose Epitom when the team needs multi-engine scheduled measurement, trends, competitor comparison, and cited sources with concrete fixes. Manual checks fit occasional checks on one or two prompts when AI search is not yet a staffed channel. ### Epitom vs. Profound URL: https://tryepitom.com/compare/epitom-vs-profound Verdict: Profound is a broad, enterprise-focused AI visibility suite with wide engine coverage and agentic-traffic analytics. Epitom pairs measurement with a technical AEO audit that generates files to deploy. The choice is engine breadth and traffic analytics versus a shorter path from insight to technical fix. The comparison describes Profound as an enterprise GenAI marketing platform with Answer Engine Insights, Agent Analytics, Prompt Volumes, Shopping, and no-code Agents. It describes Epitom as covering the main buyer-facing answer surfaces, scoring six metrics, classifying answers by persona and funnel stage, and generating deployable llms.txt and robots.txt outputs. | Capability | Epitom | Profound | | --- | --- | --- | | Positioning | Measurement plus technical AEO audit that generates files | Broad enterprise AI marketing suite | | Engines | ChatGPT, Perplexity, Gemini, Claude, Google AI Overviews | Around 8–9, including Grok, Copilot, Meta AI, DeepSeek | | Metrics | Six, including depth, position, and competitive context | Visibility, SOV, sentiment, and citations | | Persona view | Buyer persona and funnel-stage classification | Not a stated focus in this comparison | | Technical output | Deployable llms.txt and robots.txt | Content tooling and Agents | | Crawler data | Robots-rule crawlability analysis | Agent Analytics on real crawler visits | | Automation | Workflow builder with about 70 steps plus MCP | No-code Agents | | Stage | Earlier and access-gated | Established enterprise vendor with public customer stories | Choose Epitom for measurement tied to the files and fixes that earn citations, deeper per-answer scoring, persona views, and MCP-connected workflows. Choose Profound when the priority is the widest engine coverage, agentic-traffic analytics, demand data, or an established enterprise vendor. ### Epitom vs. Peec AI URL: https://tryepitom.com/compare/epitom-vs-peec-ai Verdict: Peec AI is a clean, well-adopted AI search analytics tool for marketing teams and agencies. Epitom adds a technical AEO audit and deeper answer scoring. The choice is a proven monitoring workflow versus a tighter loop from measurement to on-page fix. The comparison describes Peec as tracking visibility, position, sentiment, sources, competitors, prompts, and search volumes, with multi-country tracking and integrations such as Looker Studio, API, MCP, and SSO. It describes Epitom as adding Depth, Competitive Context, persona-level views, technical audits, deployable files, workflow integrations, and MCP. | Capability | Epitom | Peec AI | | --- | --- | --- | | Engines | ChatGPT, Perplexity, Gemini, Claude, Google AI Overviews | ChatGPT, AI Mode, AI Overviews, Copilot, Perplexity, Gemini | | Claude access | Included in Epitom’s current comparison | Enterprise tier, according to cited third-party reviews | | Metrics | Six, including depth and competitive context | Visibility, position, and sentiment | | Persona view | Buyer persona and funnel-stage classification | Not a stated focus in this comparison | | Source analysis | Cited domains per answer | Detailed source view including G2, Reddit, LinkedIn, and news | | Technical output | Deployable llms.txt and robots.txt | Actions and recommendations | | Traffic attribution | Via workflow steps such as GSC and GA4 | Built-in visibility-to-traffic attribution | | Adoption | Earlier and access-gated | Thousands of teams and 4.9/5 on G2 as described on the page | Choose Epitom for depth and competitive-context scoring, technical outputs, persona-level views, and workflows. Choose Peec for an established monitoring tool, detailed source analysis, built-in traffic attribution, Copilot or Google AI Mode tracking, and broad agency adoption. --- ## 9. Dashboard Metrics Guide URL: https://tryepitom.com/docs/metrics The guide explains how Epitom asks buyer-style questions to Google AI Mode, ChatGPT, Claude, Perplexity, Gemini, and other engines, then scores each answer for brand presence, framing, placement, and sources. Failed calls are ignored; only successful answers count. Dashboard locations: - **Overview**: Brand metrics, trends, and top sources. - **Prompts**: Per-prompt metrics, Mentions, and Volume. - **Domains / URLs**: Cited sources and Gap Score. - **Crawlability**: AI-bot access. - **Insights**: Findings from the metrics above. The guide contains 12 sections: 1. **Visibility — 0–100%**: `(answers mentioning brand ÷ total successful answers) × 100`; measures presence inside the answer, not a Google rank. 2. **Share of Voice — 0–100%**: `brand mentions ÷ (brand mentions + competitor mentions)`; shows whether a visible brand is dominant or crowded. 3. **Sentiment — 0–100**: average of positive 100, neutral 50, and negative 0; “-” means no mentions. 4. **Position — 0.0–1.0**: average weight of top 1.0, middle 0.6, bottom 0.3, and absent 0.0. 5. **Prompt-level metrics**: Visibility, SOV, Sentiment, Position, Mentions, and a seven-day Volume sparkline scoped to one question. 6. **Retrieval Rate — 0–100%**: `(answers citing source ÷ total successful answers) × 100`; distinguishes a domain from an exact URL. 7. **Citation Rate (domains) — 0.0+**: how many pages from a domain appear in one answer when the domain is used; above 1.0 means multiple pages. 8. **Competitor Citations and Brand Mentions — counts**: contested sources and URL citations that also name the tracked brand. 9. **Gap Score — 0–100**: owned-site headroom, other-domain opportunity, and other-URL opportunity adjusted for existing brand co-mentions. 10. **Crawlability — Allow Rate %**: known AI/search-bot access according to robots.txt, shown as Allowed, Partial, or Blocked. 11. **Quick cheat sheet**: maps metric patterns to focuses such as cite-worthy content, comparison content, richer category content, messaging review, or unblocking bots. 12. **Scales at a glance**: higher means more presence for Visibility, more brand talk for SOV, more positive framing for Sentiment, earlier placement for Position, more source use for Retrieval Rate, more opportunity for Gap Score, and more bots allowed for Allow Rate. --- ## 10. AEO, AI-search, and Schema.org glossary URL: https://tryepitom.com/glossary ### Core AEO and GEO concepts - **Answer Engine Optimization (AEO)**: The practice of getting a brand cited and recommended inside AI-generated answers, rather than ranking on a traditional search results page. - **Generative Engine Optimization (GEO)**: The discipline of optimizing content and brand signals so generative AI engines surface, mention, and cite a brand when composing an answer. - **Search Engine Optimization (SEO)**: The practice of improving traditional search ranking; AEO and GEO extend the same goal to the AI answer layer. - **Answer engine**: A system that responds with a direct, synthesized answer, often citing a few sources, instead of only a ranked list of links. - **Generative engine**: An AI system that composes responses from a large language model, training data, and increasingly live web retrieval. - **Search engine results page (SERP)**: A page of ranked links and features returned for a query, increasingly blended with AI Overviews. - **Zero-click search**: A query resolved directly on the results surface through an AI Overview or answer box without a click-through. - **AI citation**: The reference an answer engine gives to a source used in a generated answer. - **Grounding**: Tying a model’s answer to retrieved, verifiable sources. - **Retrieval-Augmented Generation (RAG)**: Fetching relevant documents at query time and feeding them to a language model so the answer can be grounded. - **Hallucination**: A confident but false or unsupported AI statement. - **Brand mention**: Any company or product reference inside an AI-generated answer, whether or not it links back. - **Prompt**: The natural-language question or instruction given to an AI engine. - **Query fan-out**: Expanding one user question into related sub-queries, retrieving sources for each, and synthesizing one answer. - **Passage (chunk)**: A self-contained block of text an answer engine can retrieve and quote independently. - **Passage-level citability**: How easily a paragraph can be lifted and quoted on its own when it states subject, context, and answer together. - **Large language model (LLM)**: An AI system trained on large volumes of text to predict and generate language. - **Entity**: A distinct recognizable company, product, person, or concept resolved by an engine. - **Knowledge graph**: A structured network of entities and relationships used to understand and describe brands. - **Semantic search**: Search that matches by meaning rather than exact keyword overlap. - **Vector embedding**: A numerical representation of text that places similar meanings close together for retrieval. - **Prompt tracking**: Running a fixed set of prompts against AI engines on a schedule to monitor brand mentions. - **Share of model**: The extent to which a model associates a brand with its category from training data. - **Answer surface**: Any place a generated response appears, including an AI Overview, chatbot reply, or assistant answer. - **Conversational search**: Multi-turn natural-language search in which users refine questions through a dialogue. - **LLM SEO**: An informal umbrella term for optimizing a brand’s presence inside LLM answers; it overlaps with AEO and GEO. ### Epitom’s six visibility metrics - **GEO Visibility**: How often a brand appears in AI-generated responses across engines and query types. - **Share of Voice**: The percentage of category brand mentions in AI answers that belong to the tracked brand versus competitors. - **Depth**: The richness of the brand’s presence, from a passing reference to a detailed recommendation. - **Sentiment**: The tone AI engines use about a brand: recommended, neutral, or cautioned against. - **Position**: Where the brand appears within the answer; earlier placement is generally more prominent. - **Competitive Context**: How the brand is positioned relative to others answering the same query. ### AI engines and answer surfaces - **ChatGPT**: OpenAI’s conversational AI assistant, with web search capable of retrieving and citing live sources. - **Google Gemini**: Google’s family of AI models and consumer assistant. - **Google AI Overviews**: AI-generated summaries at the top of Google results that cite selected sources. - **Google AI Mode**: A conversational, AI-first Google search experience for complex multi-part queries. - **Perplexity**: An AI answer engine that synthesizes answers with inline citations. - **Microsoft Copilot**: Microsoft’s assistant, connected to the Bing index and cited answer surfaces. - **Claude**: Anthropic’s conversational AI assistant, with web access capable of retrieving and referencing current sources. - **Grok**: xAI’s conversational assistant, integrated with X and capable of using real-time posts and web results. - **Meta AI**: Meta’s assistant embedded across Facebook, Instagram, WhatsApp, and Messenger. - **Google Search**: Traditional search increasingly blending classic results with AI Overviews and AI Mode. ### AI crawlers and bots - **GPTBot**: OpenAI’s crawler for publicly available web content used to help train models; robots.txt controls access. - **OAI-SearchBot**: OpenAI’s crawler for the search index that surfaces and links sites in ChatGPT answers. - **ChatGPT-User**: An OpenAI agent that fetches a specific page in real time when a user request requires it. - **ClaudeBot**: Anthropic’s web crawler for public content used to help train and improve Claude. - **anthropic-ai**: A user-agent token associated with Anthropic web access. - **PerplexityBot**: Perplexity’s crawler for indexing pages that can be retrieved and cited in answers. - **Google-Extended**: A robots.txt token controlling whether content is used to train and ground Google Gemini models independently of normal Search crawling. - **Applebot-Extended**: An Applebot control for whether crawled content is used to train Apple’s generative models. - **Bingbot**: Microsoft’s search crawler for the Bing index that feeds Microsoft Copilot. - **CCBot**: Common Crawl’s crawler for an open web dataset used by AI systems. ### AEO techniques and signals - **llms.txt**: A proposed root-level plain-text file containing a curated Markdown summary of a site’s facts and priority pages. - **robots.txt (AI directives)**: A root file that tells crawlers, including AI user agents, which site areas they may access. - **Structured data (JSON-LD)**: Machine-readable Schema.org markup that labels page entities and facts for engines. - **FAQPage schema**: Structured markup pairing questions with accepted answers. - **DefinedTerm schema**: Schema.org markup for a glossary term and definition within a DefinedTermSet. - **Comparison pages**: “X vs Y” or alternatives pages answering build-versus-buy questions directly. - **Newsroom**: A brand-owned hub of product news, launches, and milestones that supplies fresh citable material. - **Glossary strategy**: Publishing clear, structured definitions so engines can cite the brand when terms arise. - **E-E-A-T**: Experience, Expertise, Authoritativeness, and Trustworthiness, the quality signals supporting trusted content. - **Content freshness**: How recently a page was published or updated. - **Heading hierarchy**: Logical use of one H1 and descending H2/H3 headings so engines can extract page structure. - **Internal linking**: Connecting related pages with descriptive anchors so crawlers can discover and understand important content. - **Prerendering (SSR for agents)**: Delivering fully formed HTML so crawlers that do not execute JavaScript can read page content. - **Canonical URL**: The tag that identifies the authoritative version of a page and reduces duplicate-content confusion. ### Tools and platforms - **AI visibility platform (AEO tool)**: Software that measures how often and how favorably a brand appears across AI answer engines and helps teams act on gaps. - **Epitom**: A full-stack AEO platform that tracks brand visibility across 12+ AI answer-engine variants, scores six proprietary metrics, and turns gaps into an AEO action plan. - **Profound**: An AI visibility platform that helps brands track and analyze mentions across AI answer engines. - **Peec AI**: An AI visibility tool for monitoring mentions and share of voice across AI answer engines over time. - **Otterly.AI**: An AI search monitoring tool for brand mentions, links, sentiment, and AI Overviews. - **Scrunch AI**: A platform for monitoring and optimizing brand presence across AI search and answer engines. - **Ahrefs Brand Radar**: An Ahrefs feature that tracks brand mentions across AI Overviews and AI assistant answers. - **Semrush AI Toolkit**: A Semrush offering for brand visibility and sentiment across AI answer engines. --- ## 11. Structured data and machine-readable site signals The Epitom marketing site publishes JSON-LD that reflects the page being rendered. The site-wide graph identifies: - **Organization**: Epitom, legal name V AND N AI Private Limited, corporate URL https://www.vandn.ai, product URL https://tryepitom.com, logo, brand, and AEO-related areas of knowledge. - **WebSite**: The tryepitom.com website, published by the Epitom organization and about the Epitom software product. Page-specific Schema.org entities include: - the homepage’s WebPage, SoftwareApplication, Offer, OfferCatalog, and FAQPage; - each free tool’s WebPage, WebApplication, FAQPage, Offer, and BreadcrumbList; - tools and comparison hubs’ CollectionPage, ItemList, and BreadcrumbList; - comparison articles’ WebPage, Article, FAQPage, and BreadcrumbList; - the glossary’s WebPage, DefinedTermSet, and DefinedTerm entities; - the metrics guide’s WebPage, TechArticle, and BreadcrumbList; - Epitom Signal’s industry-report directory CollectionPage and ItemList; - the blog index’s CollectionPage and dynamic BlogPosting ItemList; and - individual blog posts’ BlogPosting and BreadcrumbList entities. The site’s per-route head manager sets the title, description, robots directive, canonical URL, Open Graph fields, Twitter fields, and page-scoped JSON-LD. Production builds prerender indexable routes so crawlers receive page content and route-specific metadata without needing to execute the React application. --- ## 12. Canonical page map - **Home**: https://tryepitom.com/ - **Free AI Visibility Audit**: https://tryepitom.com/free-audit - **Free tools hub**: https://tryepitom.com/tools - **AI Crawler Checker**: https://tryepitom.com/tools/ai-crawler-checker - **llms.txt Generator**: https://tryepitom.com/tools/llms-txt-generator - **Comparison hub**: https://tryepitom.com/compare - **Manual ChatGPT comparison**: https://tryepitom.com/compare/epitom-vs-manual-chatgpt - **Profound comparison**: https://tryepitom.com/compare/epitom-vs-profound - **Peec AI comparison**: https://tryepitom.com/compare/epitom-vs-peec-ai - **Dashboard Metrics Guide**: https://tryepitom.com/docs/metrics - **AEO and AI-search glossary**: https://tryepitom.com/glossary - **Blog index**: https://tryepitom.com/blogs - **Contact sales**: https://tryepitom.com/sales - **Privacy Policy**: https://tryepitom.com/privacy - **Terms of Service**: https://tryepitom.com/terms Research remains a coming-soon page. Epitom Signal now has a public report directory at https://tryepitom.com/signal; individual embedded report routes may remain noindex while their long-form report metadata is refined. --- ## 13. Company and contact - **Legal entity**: V AND N AI Private Limited. - **Corporate website**: https://www.vandn.ai. - **Product website**: https://tryepitom.com. - **Sales and product walkthroughs**: https://tryepitom.com/sales. - **Public crawler policy**: Epitom’s robots.txt allows general crawling and explicitly welcomes major AI and search crawlers; see https://tryepitom.com/robots.txt.