# Zumi: Full Reference > Zumi is an AI Search Intelligence Platform for marketing leaders, CMOs, Heads of Marketing, and Chief Marketing Officers at mid-to-large brands and agencies. It tracks where a brand appears, and where it disappears, across the AI engines buyers use to discover and evaluate products. This file is a fuller companion to [/llms.txt](https://www.zumihq.com/llms.txt): the same map, plus complete pricing, engine, and article detail for deeper machine reading. ## What Zumi does Zumi tracks, analyzes, and optimizes how brands appear across the AI engines buyers use to discover and evaluate products: ChatGPT, Gemini, Perplexity, Claude, Copilot, Grok, Google AI Overviews, Google AI Mode, DeepSeek. Zumi serves in-house brand teams and agencies, measuring AI Visibility and Citations. Engine coverage is a platform capability: up to 9 engines, with the tracked set scoped to each plan and requirement. It is not "all 9 on every engagement." ## AI Visibility AI Visibility describes how prominently a brand appears across tracked AI engines, refreshed daily. It is composed of four descriptive signals: - Mention rate: how often a brand is mentioned in response to tracked queries. - Share of voice: how often a brand appears relative to competitors on the same queries. - Average position: where a brand lands among the brands mentioned in an answer. - Citation share: how often a brand's own pages are cited as a source. Signals are normalized per engine, then normalized across engines by category traffic share. ## SKU-level AI visibility Brand AI visibility measures whether a brand appears in category answers. SKU-level AI visibility measures whether individual products appear in answers to buyer questions. Zumi includes both on every plan. AI visibility tracking down to the individual product or SKU, distinct from the category and segment growth rollups in E-commerce tracking, so a retail brand sees which specific products AI engines mention and which they skip. Average position is reported at the individual product or SKU level in tracked AI answers. ### Automatic product discovery and name handling Products mentioned in tracked AI answers are detected automatically and surfaced for review, each classified as a product of a brand or as another name for the brand. Products across the market can be tracked when they appear in those answers, beyond preselected brands and competitors. Tracking does not require a product catalog or website integration. Alternate names for a product, including variations in size, weight, volume, or packs, can be included under one tracked product. Similar names for the same brand can also be merged. SKU tracking and ecommerce category and segment growth tracking are included on Starter, Growth, and Enterprise. Category and segment views show trends in AI answers alongside individual-product visibility. Tracked AI engines are scoped to the plan and requirement; engine coverage alone does not specify coverage of a dedicated shopping surface. Source: [SKU-level AI visibility tracking](https://www.zumihq.com/solutions/product). ## Agency workspaces, reporting and access Multiple client workspaces and white-label reports and dashboards are included on every agency tier. Enterprise includes unlimited agency workspaces. Brand plans have a single workspace and do not include white-label reporting. - White-label reporting: Reports and dashboards rebranded to the agency, with no Zumi branding, to share with clients. - Data export, every plan: Underlying measurement data can be exported from the platform rather than being readable only inside the dashboard. - Workspace seats, every plan: No per-seat limit. Every plan allows unlimited users in the workspace. - API access, Enterprise: Programmatic access to Zumi data for integration into other systems. - Role-based access control, every plan: Granular permissions for who can view, edit, or export data within the account. - Model Context Protocol (MCP) access, Enterprise: Model Context Protocol access so AI agents and tools can query Zumi data directly. - Recommendations, every plan within monthly caps: Prioritized, per-engine actions that show what to fix to improve AI visibility, structured as Owned Media (content to create) and Earned Media (coverage to earn). Zumi measures the discovery layer: presence in AI answers and cited sources. Visibility reporting does not establish click, conversion, or revenue attribution. The available integrations do not turn a visibility measurement into a revenue measurement. Sources: [Agency capabilities](https://www.zumihq.com/solutions/agencies), [Plan scope and pricing](https://www.zumihq.com/pricing). ## Product pillars Zumi's two product pillars are AI Visibility and Citations. ## Pricing Public pricing, three tiers: - **Starter**: $99/mo, 2 engines (ChatGPT, Google AI Overviews), 50 tracked prompts. - **Growth**: $399/mo, 3 engines (adds Perplexity), 100 tracked prompts. - **Enterprise**: custom, up to 9 engines, scoped and priced to the footprint in a demo. No published Enterprise figure. Recommendations, a website audit, and every integration (Google Analytics, Google Search Console, Looker Studio, Bing Webmaster Tools) are included on every plan, including Starter. Recommendation and audit caps by plan: - Recommendations: Starter 8/mo, Growth 20/mo, Enterprise unlimited. - Website audits: Starter 1/mo, Growth 4/mo, Enterprise unlimited. Enterprise-only features: API and MCP access. SKU-level tracking, ecommerce category and segment growth tracking, data export, and role-based access control are included on every plan. White-label reporting is included on all agency tiers (Starter, Growth, Enterprise), not on brand plans. Zumi supports brands and agencies in any industry; product tracking applies where there are products to track. ## Supported AI engines Zumi covers up to 9 AI engines. The tracked set is scoped to the plan and requirement: 1. ChatGPT 2. Gemini 3. Perplexity 4. Claude 5. Copilot 6. Grok 7. Google AI Overviews 8. Google AI Mode 9. DeepSeek ## Pages - [Platform Overview](https://www.zumihq.com/platform): Zumi is an AI search intelligence platform that tracks brand presence in one dashboard, with AI Visibility tracking across up to 9 AI engines, scoped to the requirement. - [AI Search Visibility](https://www.zumihq.com/platform/ai-search): Brand presence tracked across up to 9 AI engines, scoped to the requirement. Includes AI Visibility (mention rate, share of voice, average position, and citation share), competitor share of voice, citation source analysis, query-level breakdown, optimization recommendations, and daily trends. - [How It Works](https://www.zumihq.com/platform/how-it-works): Four-step process: Define (set tracking scope and competitors), Monitor (daily queries across all tracked engines), Analyze (AI Visibility and engine breakdown), Optimize (prioritized action list ranked by impact). - [Pricing](https://www.zumihq.com/pricing): Public pricing as detailed above. - [Solutions](https://www.zumihq.com/solutions): Five use-case entry points into Zumi, for both in-house brands and agencies: AI Visibility, Content Creation, Digital PR, Agencies, and Product Visibility. - [AI Visibility](https://www.zumihq.com/solutions/ai-visibility): The core pillar. Mention rate, share of voice, average position, and citation share across tracked engines, refreshed daily, for CMO and board reporting. - [Content Creation](https://www.zumihq.com/solutions/content): Owned-media intelligence. The content gaps and cited pages that inform marketing priorities. Zumi directs what to create; it does not write the content. - [Digital PR](https://www.zumihq.com/solutions/digital-pr): Earned-media intelligence. The third-party sources AI engines cite for a category and where the brand is missing. Zumi shows where to earn coverage; it does not run outreach. - [Agencies](https://www.zumihq.com/solutions/agencies): Multi-client AI visibility with SKU tracking and white-label reporting on every agency tier. Enterprise adds API and MCP access and unlimited agency workspaces. - [SKU-level AI Visibility](https://www.zumihq.com/solutions/product): Individual-product tracking for retail brands, ecommerce teams, and agencies on every plan. Brand tracking and SKU tracking are different measurement levels. Ecommerce category and segment growth tracking is also included on every plan. - [About](https://www.zumihq.com/about): Zumi's mission: help marketing leaders understand and improve where their brand appears across the AI engines where buyers discover and evaluate brands. - [Book a Demo](https://www.zumihq.com/book-demo): Personalized demo, including live visibility data for the prospect's brand. - [Talk to Sales](https://www.zumihq.com/talk-to-sales): Discuss the brand, category, and tracking requirements. - [Contact](https://www.zumihq.com/contact): Sales, support, and press inquiries. - [Privacy Policy](https://www.zumihq.com/privacy): Data collection, use, and privacy rights. - [Terms of Service](https://www.zumihq.com/terms): Terms governing use of the website and platform. ## Resources / Blog Guides, research, and playbooks on AI visibility, GEO (Generative Engine Optimization), and search visibility strategy. Full index: https://www.zumihq.com/resources - [Citation analysis tools for AI search: what to check](https://www.zumihq.com/resources/citation-analysis-tools-for-ai-search): Citation analysis tools for AI search should list sources by engine, answer, and owner. What to check, and what citations can't prove. Read the guide. (Guide, 2026-10-10) · [Markdown](https://www.zumihq.com/resources/citation-analysis-tools-for-ai-search.md) - [How to compare AI search tracking tools](https://www.zumihq.com/resources/how-to-compare-ai-search-tracking-tools): A two-day trial for comparing AI search tracking tools on the same questions, with the day-to-day change that counts as normal. Read the guide. (Guide, 2026-10-10) · [Markdown](https://www.zumihq.com/resources/how-to-compare-ai-search-tracking-tools.md) - [How to track a brand in AI Overviews and AI Mode](https://www.zumihq.com/resources/track-brand-in-ai-overviews-and-ai-mode): AI Overviews and AI Mode cite different sources and change daily. How to set up tracking that holds up, based on 77 tracked Google answers. (Playbook, 2026-10-10) · [Markdown](https://www.zumihq.com/resources/track-brand-in-ai-overviews-and-ai-mode.md) - [Product visibility in AI answers: why counts break](https://www.zumihq.com/resources/track-product-visibility-in-ai-answers): Product visibility in AI answers breaks in five ways, from one product split across names to retailer pages. How to fix the count. Read the guide. (Guide, 2026-10-10) · [Markdown](https://www.zumihq.com/resources/track-product-visibility-in-ai-answers.md) - [AEO vs GEO: which term engines and buyers use](https://www.zumihq.com/resources/aeo-vs-geo): AEO and GEO name the same work. Zumi data shows which term buyers search for worldwide and in four markets, and which AI engines use. (Guide, 2026-10-08) · [Markdown](https://www.zumihq.com/resources/aeo-vs-geo.md) - [For One India Beauty Buyer, AI Engines Agree on the Source Before the Brand](https://www.zumihq.com/resources/india-beauty-cross-engine-agreement-gap): Across 54 India beauty search queries, 7 of 8 buyer segments show AI engines agreeing more on brands than sources. One segment inverts it. (Research, 2026-08-25) · [Markdown](https://www.zumihq.com/resources/india-beauty-cross-engine-agreement-gap.md) - [The dark funnel in AI search, measured](https://www.zumihq.com/resources/dark-funnel-ai-search): Across 77 tracked answers, Google AI Overviews used the term dark funnel in 8 of its 25. Gemini used it twice, AI Mode once. Read the research. (Research, 2026-08-23) · [Markdown](https://www.zumihq.com/resources/dark-funnel-ai-search.md) - [Gemini, AI Overviews, and AI Mode cite differently](https://www.zumihq.com/resources/gemini-ai-overviews-ai-mode-citation-differences): Zumi asked one question across Gemini, AI Overviews, and AI Mode for 19 days. The three engines cited 362 domains and shared 9. Read the study. (Research, 2026-08-17) · [Markdown](https://www.zumihq.com/resources/gemini-ai-overviews-ai-mode-citation-differences.md) - [How to evaluate AI visibility tools as answers shift](https://www.zumihq.com/resources/how-to-evaluate-ai-visibility-tools): AI answers change daily. Zumi data shows how much, and what that means when choosing AI visibility, AEO, or GEO tools. Read the guide. (Guide, 2026-07-16) · [Markdown](https://www.zumihq.com/resources/how-to-evaluate-ai-visibility-tools.md) - [AI Search Intelligence metrics that matter](https://www.zumihq.com/resources/geo-metrics-beyond-mention-rate): Four AI visibility signals answer four different questions. Learn how to read each one without hiding the result in a composite. Read the guide. (Guide, 2026-06-16) · [Markdown](https://www.zumihq.com/resources/geo-metrics-beyond-mention-rate.md) - [How to write content that AI engines cite](https://www.zumihq.com/resources/how-to-write-content-ai-engines-cite): AI citations concentrate on a few dozen domains. Zumi data shows how few, and what that means for writing content AI engines cite. Read the guide. (Playbook, 2026-06-16) · [Markdown](https://www.zumihq.com/resources/how-to-write-content-ai-engines-cite.md) - [Branded search in the AI era: the brand narrative](https://www.zumihq.com/resources/branded-search-in-the-ai-era): Branded search was the safest territory in digital marketing. AI answers have changed that, and most brands have not noticed. Read the guide. (Strategy, 2026-06-15) · [Markdown](https://www.zumihq.com/resources/branded-search-in-the-ai-era.md) - [How to audit competitor AI visibility](https://www.zumihq.com/resources/competitor-ai-visibility-audit): How to run an AI visibility audit on competitors and map which questions each rival owns across AI engines. Read the guide. (Playbook, 2026-06-14) · [Markdown](https://www.zumihq.com/resources/competitor-ai-visibility-audit.md) - [GEO for SaaS: The category ownership playbook](https://www.zumihq.com/resources/geo-for-saas): How SaaS companies establish and defend category ownership in AI answers, and why challengers can win it faster than expected. Read the guide. (Playbook, 2026-06-14) · [Markdown](https://www.zumihq.com/resources/geo-for-saas.md) - [How B2B Buyers Use AI When Researching Vendors](https://www.zumihq.com/resources/b2b-ai-buyer-journey): The verified research on B2B AI adoption in purchasing, and what it means for how and when brands need to appear in AI answers. Read the research. (Research, 2026-06-13) · [Markdown](https://www.zumihq.com/resources/b2b-ai-buyer-journey.md) - [Entity Optimization: The AI Visibility Foundation](https://www.zumihq.com/resources/entity-optimization-for-ai): AI engines are entity-first systems. Before recommending a brand, an engine needs to recognize it as a coherent, well-attested entity. Read the guide. (Playbook, 2026-06-13) · [Markdown](https://www.zumihq.com/resources/entity-optimization-for-ai.md) - [GEO vs. SEO: Differences and How to Allocate](https://www.zumihq.com/resources/geo-vs-seo): GEO and SEO reward different signals, not different degrees of the same one. What diverges, what carries over, and how to allocate budget. (Guide, 2026-06-12) · [Markdown](https://www.zumihq.com/resources/geo-vs-seo.md) - [How Perplexity decides which sources to cite](https://www.zumihq.com/resources/how-perplexity-cites-sources): Perplexity runs the strictest quality gate of any major AI engine. Here is what determines whether a brand's content makes the cut. Read the guide. (Research, 2026-06-12) · [Markdown](https://www.zumihq.com/resources/how-perplexity-cites-sources.md) - [How to get a brand mentioned in ChatGPT answers](https://www.zumihq.com/resources/how-to-appear-in-chatgpt-answers): SEO for ChatGPT works differently from Google SEO. The four signals that decide which brands get mentioned, and how to track the result. (Playbook, 2026-06-12) · [Markdown](https://www.zumihq.com/resources/how-to-appear-in-chatgpt-answers.md) - [How to baseline AI visibility before buying a tool](https://www.zumihq.com/resources/baseline-ai-visibility): A manual AI visibility audit starts with fixed questions, declared scope, raw answers, and separate signals. Build a defensible first read. (Playbook, 2026-06-11) · [Markdown](https://www.zumihq.com/resources/baseline-ai-visibility.md) - [State of AI search, October 2026: the verified data](https://www.zumihq.com/resources/state-of-ai-search-2026): Verified, sourced numbers on AI search as of October 2026: adoption is settled, B2B buyers lead, and clicks fall when AI answers appear. (Research, 2026-06-11) · [Markdown](https://www.zumihq.com/resources/state-of-ai-search-2026.md) - [AI visibility is not the product](https://www.zumihq.com/resources/ai-visibility-is-not-the-product): Measurement-only GEO tools stall the moment the number arrives. The durable value is the loop: diagnose, act, then see it improve. (Strategy, 2026-06-04) · [Markdown](https://www.zumihq.com/resources/ai-visibility-is-not-the-product.md) - [What is Generative Engine Optimization (GEO)?](https://www.zumihq.com/resources/what-is-geo): A complete introduction to Generative Engine Optimization: what GEO is, why it matters, and how it differs from traditional SEO. Read the guide. (Guide, 2026-03-01) · [Markdown](https://www.zumihq.com/resources/what-is-geo.md) - [How AI engines decide what brands to mention](https://www.zumihq.com/resources/how-ai-engines-rank-brands): What published research says about the signals behind brand mentions in ChatGPT, Google's AI features, Perplexity, and Claude. Read the research. (Research, 2026-02-14) · [Markdown](https://www.zumihq.com/resources/how-ai-engines-rank-brands.md) - [The CMO guide to AI Search Intelligence](https://www.zumihq.com/resources/cmo-guide-to-ai-visibility): A 90-day operating model helps CMOs measure AI discovery, assign action, and report clear limits without inventing causal ROI. Read the guide. (Strategy, 2026-02-01) · [Markdown](https://www.zumihq.com/resources/cmo-guide-to-ai-visibility.md) - [AI engine comparison: which drives purchase intent](https://www.zumihq.com/resources/ai-engine-comparison): What public market-share and conversion data says about which AI engines matter most for which buyer types, and how to prioritize. Read the guide. (Research, 2026-01-05) · [Markdown](https://www.zumihq.com/resources/ai-engine-comparison.md) - [AI visibility platform for agencies managing multiple client brands](https://www.zumihq.com/resources/agency-guide-to-geo): What agencies should look for in an AI visibility platform for tracking, reporting on, and pricing AI visibility work across a full client roster. (Guide, 2025-12-20) · [Markdown](https://www.zumihq.com/resources/agency-guide-to-geo.md) ## Key concepts - **AI Search Intelligence Platform**: Zumi's category. A platform that measures how brands appear across the tracked AI engines, up to nine and scoped to each engagement, traces each answer to the sources behind it, and prescribes what to fix. See the [platform overview](https://www.zumihq.com/platform). - **AI Visibility**: How prominently a brand appears across all tracked AI engines and queries, measured by mention rate, share of voice, average position, and citation share. Updated daily. - **GEO (Generative Engine Optimization)**: The discipline of optimizing brand presence in AI-generated answers, analogous to SEO for traditional search engines. - **Citation Analysis**: Identifying which content, pages, and domains AI engines cite when generating answers about a category or brand. - **Share of Voice (AI)**: How often a brand leads the AI answers in which any tracked brand appears. ## Target audience Zumi serves two primary personas: - **CMOs and marketing leaders at mid-to-large brands**: need clear AI Visibility reporting and a prioritized action list for board reporting and cross-channel visibility strategy. - **Marketing agencies**: managing multiple client brands who need multi-client dashboards, white-label reporting, and scalable AI visibility as a client service.