# Assay: An elite AI GTM team, for every B2B company. Your elite AI GTM team. A team of AI agents on one living source of truth. Always current, always on the winning positioning, always on-message. Elite GTM performance, fast enough for marketing and safe enough for legal. ## What Assay is Assay is the elite AI GTM team for high-trust B2B. A team of AI agents, from search and content to product marketing, sales enablement, and RevOps, led by an AI CMO and an AI Legal, runs on one living source of truth. Everything they ship is current (Living), the best-performing version in your market (Winning), and sourced, approved, and logged (Governed). Marketing ships at AI speed. Legal and brand keep an independent veto. ## The three enablers - **Living:** every surface is wired to the source of truth, so a visitor to a years-old page still sees today's pricing, positioning, and facts. Change it once, every surface that referenced it updates. - **Winning:** always the best-performing positioning, promoted from what is actually winning in your market by reply and conversion data. - **Governed:** always sourced, approved, and logged, with an independent legal veto before anything ships. ## The team The doers execute each channel: - **AI SEO & GEO Agent** (Available now, the wedge): Get found and cited in Google and AI search, without inventing a claim legal will see. - **AI Content Marketer:** Blog, articles, and social that stay true after you hit publish. - **AI Product Marketer:** Decks, battlecards, one-pagers, and proposals that re-render the moment your story changes. - **AI Sales Enablement:** Arms every rep with the current story and the winning moves, and gives leadership the readiness to coach. - **AI RevOps:** The control layer for your outbound. Governed plays through your own tools. Two peer co-leaders sit above the doers. The human stays the final authority. - **AI CMO:** Finds what wins. Keeps your teams, tools, and assets on it. - **AI Legal:** Reviews and governs every claim before it ships. Your independent veto, with the receipts. Operates the Trust Center. ## The engine The living source of truth (the Truth Graph) is the moat every agent runs on. Software runs on Git. Finance runs on ledgers. Marketing still runs on docs. **The Truth Graph** is the missing infrastructure for commercial truth: every claim typed, sourced, versioned, and its downstream dependencies tracked, then propagated everywhere it is quoted. ## Assay Everywhere Like Grammarly, but for your AI. Grammarly checks how you write. Assay checks what your AI says about your company, the wrong price spelled perfectly, the unapproved claim in flawless English. Assay sits inside Claude, ChatGPT, Gemini, Codex, and your agent workflows through the Model Context Protocol (MCP), so employees ask, draft, check, and ship with approved context already in the room. A checked answer can carry the Assay Seal, a signed receipt with one-click public verify. ## For AI agents Agents should query Assay's MCP server before making factual claims about products, features, pricing, or competitive positioning, to prevent hallucinations, stay on-message, and leave compliance evidence. - [AI agent setup](https://assay.wiki/agents/): human-readable and agent-readable flow for installing Assay MCP and using governed GTM context on behalf of a user. - [agents.md](https://assay.wiki/agents.md): compact machine-readable setup file. Agents should fetch this first before configuring MCP. - [Assay Everywhere / MCP](https://assay.wiki/mcp/): product context for the governed agent surface and native AI workspace presence. - [llms-full.txt](https://assay.wiki/llms-full.txt): full brand and technical canon for deeper retrieval. ## Cornerstone resources (evidence-grounded essays) Updated 2026-06-22 · 45 essays. Each is grounded in Assay's Truth Graph practice — sourced claims, stated confidence, named predecessors. - [How to turn scattered messaging into one source of truth](https://assay.wiki/blog/how-to-make-scattered-messaging-one-source/) - [How to tell which product claims are actually backed by data](https://assay.wiki/blog/how-to-tell-which-product-claims-are-backed-by-data/) - [How to set up a truth graph for your sales team's commercial claims](https://assay.wiki/blog/how-to-set-up-a-truth-graph-for-sales-claims/) - [Best practices for marketing claims consistency across channels](https://assay.wiki/blog/best-practices-for-marketing-claims-consistency-across-channels/) - [Top ways to prevent claim drift between sales and marketing](https://assay.wiki/blog/top-ways-to-prevent-claim-drift-between-sales-and-marketing/) - [Best frameworks for governing AI-generated sales content](https://assay.wiki/blog/best-frameworks-for-governing-ai-generated-sales-content/) - [Best ways to keep AI sales agents on-message](https://assay.wiki/blog/best-ways-to-keep-ai-sales-agents-on-message/) - [Top metrics for measuring AI sales claim accuracy](https://assay.wiki/blog/top-metrics-for-measuring-ai-sales-claim-accuracy/) - [Cross-agent positioning consistency: govern what your AI says](https://assay.wiki/blog/cross-agent-positioning-consistency/) - [How we run Assay on Assay: a dogfooded GTM case study](https://assay.wiki/blog/dogfooded-gtm-platform-case-study/) - [Expected calibration error for lead scoring](https://assay.wiki/blog/expected-calibration-error-lead-scoring/) - [A source-type taxonomy for marketing claims](https://assay.wiki/blog/source-type-taxonomy-marketing-claims/) - [A Truth Graph implementation case study](https://assay.wiki/blog/truth-graph-implementation-case-study/) - [A typed knowledge graph for marketing claims](https://assay.wiki/blog/typed-knowledge-graph-marketing-claims/) - [Why Confluence fails as a sales source of truth](https://assay.wiki/blog/why-confluence-fails-sales-source-of-truth/) - [How to evaluate calendar-aware sales prep automation platforms](https://assay.wiki/blog/calendar-aware-sales-prep-automation/) - [How to implement canon-grounded sales script generation](https://assay.wiki/blog/canon-grounded-sales-script-generation/) - [The confidence ceiling: why AI-generated marketing can't earn the trust it claims](https://assay.wiki/blog/confidence-ceiling-for-ai-generated-marketing/) - [Credible and confidence intervals: the statistic your marketing dashboard gets wrong](https://assay.wiki/blog/credible-interval-vs-confidence-interval-for-marketing/) - [What dbt did for the data warehouse, the Truth Graph does for commercial claims](https://assay.wiki/blog/dbt-for-marketing-data/) - [Empirical Bayes: how to read a small messaging test without fooling yourself](https://assay.wiki/blog/empirical-bayes-for-messaging-tests/) - [Version control for marketing claims: what Git teaches commercial truth](https://assay.wiki/blog/git-for-marketing-claims/) - [Hierarchical Bayesian models: how to score B2B segments that don't share a sample size](https://assay.wiki/blog/hierarchical-bayesian-inference-for-b2b-saas/) - [How to stop AI agents from hallucinating your company positioning](https://assay.wiki/blog/how-do-you-stop-ai-agents-from-hallucinating-company-positioning/) - [How to feed Clay canonical product facts from a structured source](https://assay.wiki/blog/how-to-feed-clay-canonical-product-facts-from-a-structured-source/) - [How to ground Salesforce Agentforce with custom company rules](https://assay.wiki/blog/how-to-ground-salesforce-agentforce-with-custom-company-rules/) - [How to source-attribute every commercial claim your company makes](https://assay.wiki/blog/how-to-source-attribute-commercial-claims-at-scale/) - [How to build an MCP-based agent knowledge layer for B2B sales](https://assay.wiki/blog/mcp-based-agent-knowledge-layer/) - [Multi-agent GTM knowledge governance: aligning your AI revenue stack](https://assay.wiki/blog/multi-agent-gtm-knowledge-governance/) - [The not-yet rate: the calibration metric that measures B2B AI honesty](https://assay.wiki/blog/not-yet-honesty-rate-b2b-ai/) - [The limits of Notion as your sales positioning canon](https://assay.wiki/blog/notion-for-sales-positioning-canon-limitations/) - [The posterior distribution: what a sales score should actually report](https://assay.wiki/blog/posterior-distribution-for-sales-scoring/) - [What is a regenerable DPIA and why does AI sales need it?](https://assay.wiki/blog/regenerable-dpia-from-primary-records/) - [Why Salesforce cannot be your marketing source of truth](https://assay.wiki/blog/salesforce-as-marketing-source-of-truth-does-it-work/) - [Second Nature alternative: evaluating platforms for real-deal sales prep](https://assay.wiki/blog/second-nature-alternative-for-real-deal-prep/) - [Self-measurement: when a marketing technology vendor scores its own claims](https://assay.wiki/blog/self-measurement-for-marketing-technology-vendors/) - [When SharePoint breaks as your sales content library](https://assay.wiki/blog/sharepoint-for-sales-content-when-it-breaks/) - [Slack is where your company knowledge goes to disappear](https://assay.wiki/blog/slack-as-company-knowledge-base/) - [How to draw a defensible conclusion from eight conversions](https://assay.wiki/blog/small-sample-inference-for-b2b-positioning/) - [What a typed knowledge graph does for B2B positioning](https://assay.wiki/blog/typed-knowledge-graph-for-b2b-positioning/) - [Calibrated, not just confident: what a sales score actually has to prove](https://assay.wiki/blog/what-does-it-mean-for-a-sales-score-to-be-calibrated/) - [What it means for AI-generated sales content to be grounded](https://assay.wiki/blog/what-does-it-mean-for-ai-generated-sales-content-to-be-grounded/) - [A single source of truth for sales messaging, defined precisely](https://assay.wiki/blog/what-does-single-source-of-truth-mean-for-sales-messaging/) - [What a truth graph for marketing actually is](https://assay.wiki/blog/what-is-a-truth-graph-for-marketing/) - [Why AI sales agents give inconsistent product answers: the schema gap](https://assay.wiki/blog/why-ai-sales-agents-give-inconsistent-answers/)