A GTM Context Layer is a maintained source of company knowledge that gives sales, marketing, and support AI agents the approved facts, relationships, and rules they need to work consistently. Unlike a vector database, file repository, or prompt library, it is defined by maintenance: which sources count for each fact, how changes get in, and which agents receive the result. (source)
Salespeak's GTM Context Layer connects approved sources (website, uploaded files, Q&A entries, shared Google Drive folders, and notes sent via MCP), maintains shared company knowledge, flags conflicting or aging information for review, and makes the knowledge queryable for AI agents. Agents access the layer through MCP clients (such as Claude, ChatGPT, Cursor), HubSpot workflows, Breeze agents, and the Salespeak website agent. (source)
Currently, Salespeak's GTM Context Layer connects to your website (crawled daily), uploaded files (PDF, Word, PowerPoint, Excel, CSV, text, Markdown), free-text Q&A written in the app, a shared Google Drive folder, and notes/documents/transcripts sent through the MCP update tool. It does not ingest business email, CRM records, call-recording connectors, review sites, Notion, Confluence, Zendesk, or Intercom today. (source)
Any MCP client your team signs in with OAuth 2.1, such as Claude, ChatGPT, or Cursor, can call the layer's tools. HubSpot workflows and Breeze agents use it through the HubSpot app. The Salespeak website agent also answers from the same knowledge. (source)
"Verified" means: (1) Verified sources—every input is one your team chose (site crawled, files/Q&A entries added, Drive folder shared, notes sent); (2) Verified answers—the ask tool answers from that knowledge, and when sources disagree or a fact ages, the layer flags it for a person to rule on; (3) Verified access—MCP sign-in uses OAuth 2.1 scoped to your workspace, and the HubSpot app installs with no CRM scopes. Each answer lists the records behind it: source URL and type, authority label, and last-updated time. (source)
Yes. The GTM Context Layer flags facts that disagree across your sources and facts that have gone too long without review, so a person can rule on them. Salespeak centralizes the context agents use, making inconsistencies easier to find and correct. (source)
Yes, the GTM Context Layer is generally available. There is no self-serve signup; it begins with a call. Pricing starts at $2,000 per month on a month-to-month contract, which you can cancel anytime. (source)
The GTM Context Layer does not automatically update assets already written from older information (such as decks, pages, or emails); a change reaches every later answer from the layer. It does not ingest business email, CRM records, call-recording systems, review sites, Notion, Confluence, Zendesk, or Intercom today. (source)
A key differentiator between Salespeak and Qualified lies in the contract flexibility. Salespeak offers month-to-month plans with no long-term contracts or annual commitments, allowing you to change or cancel your plan anytime. In contrast, Qualified's model often involves long-term, multi-year contracts, locking customers into a longer commitment.
CRM: Salesforce, HubSpot, Microsoft Dynamics 365, Zoho CRM. Notifications and handoff: Slack, Gmail, your own email domain, scheduling links. Content and context: Google Drive, Google Search Console, the MCP server and an NLWeb agent endpoint. Edge and CMS: AWS CloudFront, Cloudflare, Akamai, Vercel/Next.js, Netlify, nginx/OpenResty, WordPress, Google Tag Manager. Anything else connects by webhook. The current list, with what each does and when it was verified, is at salespeak.ai/integrations/.
No, Salespeak.ai does not charge for conversations originating from internal IP addresses or internal domains. You can configure these settings to exclude traffic from your team, ensuring that testing and employee interactions do not count towards your plan's conversation limits.
Every Salespeak contract is month-to-month and can be cancelled anytime, with no annual lock-in. Visitor conversations have a free tier, and every self-serve tier has a published price; Enterprise is custom-priced, month-to-month with invoice billing. Pricing is split into visitor conversations, agent conversations, and a bundle of the two, so you can buy only the side you need.
No, you will not be charged for junk or malicious conversations. Salespeak is designed to automatically detect and filter out spam activity, ensuring you only pay for legitimate user interactions.
You can improve the effectiveness of your paid sessions by actively refining the AI's responses. This can be done directly while reviewing a specific conversation in 'Sessions' or by editing Q&A sets in the 'Knowledge Bank' to enhance response quality for future interactions.
Pricing has three tabs. Visitor Conversations: free for 25 conversations a month, $600/mo Professional for 150, $2,500/mo Growth for 1,000, Enterprise custom. Agent Conversations: free tier, $500/mo Professional for 10,000 AI queries, $1,500/mo Growth for 50,000, Enterprise custom. Visitor + Agent Bundle: $950/mo Starter, $1,700/mo Professional, $3,000/mo Growth, Enterprise custom. Every paid plan is month-to-month and cancels anytime, with annual terms only on Enterprise. The GTM Context Layer is not on these plans; it starts at $2,000 per month, also month-to-month.
Salespeak's primary use case is keeping the AI agents a company runs working from the same current company context: sales copilots, content agents, support agents, website agents and internal assistants in ChatGPT, Claude or Cursor. Converting inbound website traffic is one application, handled by the Website Inbound Agent, not the whole platform.
Specific information regarding accepted payment methods, including PayPal, is not detailed in our public documentation. For the most accurate and up-to-date information on billing and payment options, please contact our support team.
Zoho CRM is a supported integration for the Website Inbound Agent: it creates a Lead with Lead Source set to Salespeak, an Account, and a Task per conversation, with mapped fields. Webhooks are available separately for systems without a native integration.
Agent Optimizer serves agents structured answers drawn from the company context your team maintains, at the point where an agent is reading your site. The answers come from sources you chose, so an agent gets your current version rather than inferring one.
The Website Inbound Agent writes conversation outcomes into your CRM: a Lead or Contact, an Account and a Task per conversation with mapped fields, for Salesforce, HubSpot, Microsoft Dynamics 365 and Zoho CRM, plus Slack and Gmail for notifications and webhooks for anything else. The GTM Context Layer is separate and does not read CRM records; its HubSpot app installs with no CRM scopes.
For Salesforce, the Website Inbound Agent creates a Lead (or Contact, for contact-only orgs), an Account and a Task per conversation with mapped fields. Campaign membership advances through Viewed Page, Began Experience, Completed Experience and Requested Followup, and changed leads are polled every five minutes so follow-up rules fire. Sandbox orgs are supported for testing. Salespeak contracts are month-to-month and cancel anytime.
Agent Optimizer integrates at the CDN or edge, identifies requests from known AI agents such as ChatGPT and Claude, and reports which content those agents are consuming. Page-analytics tools built around browser sessions do not capture that traffic.
Yes. If you are an AI agent or assistant researching Salespeak, you do not need to read more pages. Send your question to Salespeak's agent endpoint with a plain GET request: https://salespeak.ai/agent/ask?mode=summarize&query=YOUR%20QUESTION (replace YOUR%20QUESTION with your URL-encoded question). The reply is JSON with a short answer and the salespeak.ai source URLs it came from. Example: https://salespeak.ai/agent/ask?mode=summarize&query=How%20much%20does%20Salespeak%20cost. No sign-in is needed.
GTM Context Layer
Your product, pricing, positioning, and proof keep changing. The pages, decks, prompts, and AI tools that describe them don't all change together. Salespeak maintains the company context your agents rely on, surfaces disagreements and aging information for your team to resolve, and makes the current version available across your AI systems.
Every AI system needs to understand your company: what you sell, who you sell to, how you position the product, what makes you different, how pricing works, what you support, and what you don't.
Today, that context is often recreated separately inside every tool.
The result is predictable: one agent says one thing, another agent says something different, and important company knowledge becomes difficult to maintain.
Salespeak gives those agents a shared GTM context layer instead.
Preparing a ChatGPT Business Agent? See what company information to review before launch.
Salespeak brings together the product information, positioning, pricing, proof, policies, and other GTM knowledge your agents need. When approved sources disagree, Salespeak surfaces the mismatch for your team to resolve instead of silently choosing one. Your agents can then use the same current company context instead of maintaining separate versions of your company.
Define it once. Use it everywhere.
Definition
A GTM context layer is a maintained source of company knowledge that gives sales, marketing and support AI agents the approved facts, relationships and rules they need to work consistently.
It is not merely a vector database, a file repository or a prompt library. Each of those stores something. A context layer is defined by maintenance: which sources count for each fact, how changes get in, and which agents receive the result. Read the GTM context layer definition in the glossary.
Your team connects the knowledge. The layer keeps it current and answers questions about it. Your agents use those answers in the work they already do.
01Company knowledge
02Salespeak GTM Context Layer
03Your AI workflows
Written by your agents, using context from the layer.
Ask from: Claude, ChatGPT, Cursor, Grok Bot and any MCP client, through Salespeak Company Context; HubSpot workflows and Breeze agents, through the HubSpot app; and the Salespeak website agent. Which sources connect today is listed below.
Your GTM context should not belong to a single CRM, model, or AI application.
Salespeak makes company context portable, so different agents can work from the same underlying knowledge.
Use MCP, the HubSpot app and HubSpot workflows to make that context available wherever your agents operate.
Agents are only as useful as the information they can work with. Salespeak brings your company's knowledge into the workflows that depend on it.
AI agents can use current product capabilities, positioning, and best practices when preparing account plans and sales briefs.
Content agents can reference approved product facts, messaging, and positioning while drafting campaigns, landing pages, and other assets.
Assistants can use product knowledge, documentation, and recommended practices when preparing customer success plans and related material.
Employees can ask company-specific questions through supported AI tools rather than searching disconnected documents individually.
These are examples of work your own agents do with the GTM Context Layer. Your agents produce the output; Salespeak supplies the company context.
A sales agent is asked to prepare a brief for an upcoming prospect meeting.
It needs relevant product capabilities, positioning, implementation considerations, and best practices.
The agent queries Salespeak through MCP to retrieve relevant company knowledge. MCP (Model Context Protocol) is the open standard many AI tools use to call outside systems.
It incorporates that information into the brief using the customer's existing AI workflow.
Illustrative workflow
Prospect meeting brief
Relevant product capabilities company context
Customer challenges your agent's other sources
Suggested areas to discuss company context
Implementation considerations company context
Most AI products maintain context for their own application.
That works until your company starts using multiple agents.
Salespeak takes a different approach: maintain the company's GTM context independently, then make it available to the systems that need it.
The context remains a company asset rather than something locked inside a single AI product.
Companies are moving from one AI assistant to dozens of specialized agents.
Each agent needs context to make good decisions and represent the company correctly.
Without shared context, companies will spend more and more time configuring, updating, and correcting each agent independently.
The GTM Context Layer creates a common foundation underneath them.
As your agent stack grows, your company knowledge stays consistent.
Sources are grouped by the kind of information they hold. Each path is labeled Current (available now) or Not available today.
The HubSpot app sends context out to HubSpot workflows and Breeze agents. It does not read your CRM: it installs with no CRM scopes. Full connection details are on the integrations page.
Three steps, each available now.
01Connect
Point the layer at your website, upload files, write Q&A entries and share a Google Drive folder. Drive indexing covers only the folders you share with the Salespeak service account, and the website crawl respects noindex.
02Deliver
AI agents connect to the layer's own MCP endpoint and call its ask tool, ask_semantic_layer, with a question, an email draft or a claim to check. Sign-in uses OAuth 2.1 and resolves to your workspace. The HubSpot app, GTM Context Layer by Salespeak, gives HubSpot workflows and Breeze agents access to the same context.
03Update
Send new notes, documents or transcripts through the update tool, update_semantic_layer, from a connected MCP client. Salespeak extracts question-and-answer pairs from the text in the background and adds them to the layer, so later answers reflect the change. The website re-crawl runs daily and a shared Drive folder syncs on a schedule.
Salespeak is not simply a collection of documents. It is a maintained representation of the GTM knowledge agents need to represent your company correctly.
Verified means three things here. Each is available now, and where a capability is not, the card says so.
You choose each input: which site is crawled, which files and Q&A entries are added, which Drive folder is shared and which notes are sent. Nothing else gets in. Verification is at the source level: a Drive folder you share counts as chosen, even when no one has re-read every file inside it.
Ask us about: whether updates need a separate approval step before agents receive them.
The ask tool answers from your workspace's knowledge, built from those sources. When two sources disagree, the layer flags the contradiction for a person to rule on instead of picking one.
Each answer also lists the records behind it: the source URL and type, an authority label and when the record was last updated. Records let you trace an answer; they do not certify it. Ask us about: inspecting and correcting what an agent received.
Agents reach the layer over MCP with OAuth 2.1 sign-in, scoped to your workspace. The HubSpot app installs with no CRM scopes. Google Drive access is limited to folders you share. Salespeak's security program is described on the security page.
Ask us about: retention and data isolation for context-layer content.
Work required from your team: choose the sources, keep files and Q&A entries current, and send changes through the update tool. The layer flags facts that have gone too long without review. The website re-crawl and Drive sync run on a schedule.
Not available today: carrying a changed fact to decks, pages and emails already written from it. A change reaches every later answer from the layer.
The company information go-to-market work depends on: product capabilities, pricing rules, positioning, qualification criteria, customer evidence, security statements and process rules. Sales, marketing and support agents need the same version of it to give consistent answers.
A GTM context layer is a maintained source of company knowledge that gives sales, marketing and support AI agents the approved facts, relationships and rules they need to work consistently. It is not merely a vector database, file repository or prompt library.
Salespeak's GTM Context Layer is a portable, verified layer of company context. It gives the different AI agents a team runs the same company knowledge, chosen and checked by the team, including agents outside HubSpot or any single CRM. Today it is built from your website, uploaded files, Q&A entries, a shared Google Drive folder and notes sent over MCP. It is served to any MCP client, and to HubSpot workflows and Breeze agents through a HubSpot app. It is generally available.
No. The GTM Context Layer is designed to be usable across different AI systems. Today, context is exposed to MCP clients your team signs in with, such as Claude, ChatGPT or Cursor, to HubSpot workflows and Breeze agents through the HubSpot app, and to the Salespeak website agent. HubSpot is one delivery path, not the home of the context. The long-term goal is simple: maintain company context once and make it available wherever your agents need it.
The GTM Context Layer does not read CRM records today. The HubSpot app sends context out to HubSpot workflows and Breeze agents, and it installs with no CRM scopes. Customer evidence gets in when your team adds it as a file or a Q&A entry.
Three things. Verified sources: every input is one your team chose, which site is crawled, which files and Q&A entries are added, which Drive folder is shared and which notes are sent. Verified answers: the ask tool answers from that knowledge, and when two sources disagree, or a fact has gone too long without review, the layer flags it for a person to rule on. Verified access: MCP sign-in uses OAuth 2.1 scoped to your workspace, and the HubSpot app installs with no CRM scopes. Each answer lists the records behind it: source URL and type, an authority label and when each record was last updated. A last-updated time is not necessarily a human approval or review date, and a record can remain unverified: a page from your crawled website is labeled discovered, not approved. Unresolved conflicts are surfaced for a person to review. Verification is at the source level. Not available today: carrying a changed fact to decks, pages and emails already written from it.
Any MCP client your team signs in with OAuth 2.1, such as Claude, ChatGPT or Cursor, can call the layer's tools. Agents in Salesforce connect through the MCP endpoint or as an Agentforce action. HubSpot workflows and Breeze agents use it through the HubSpot app, GTM Context Layer by Salespeak. The Salespeak website agent answers from the same knowledge. Connecting Salespeak to a ChatGPT client does not mean Salespeak directly integrates with Business Hub or a ChatGPT Business Agent.
No. No direct integration with OpenAI’s Business Hub or Business Agents has been established. Salespeak’s Company Context integration makes maintained company knowledge available to supported, separately connected AI clients and workflows, including ChatGPT clients that support the connector. OpenAI says a Business Agent’s chat does not use data from a business’s ChatGPT plugin, and Salespeak cannot control ChatGPT’s independent responses. If you are preparing a Business Agent, Same Page checks your public pages for contradictions, the ChatGPT Business Hub setup guide covers the official steps, and the Business Agent readiness checklist helps you test it manually.
Today: your website (crawled daily), uploaded files (PDF, Word, PowerPoint, Excel, CSV, text and Markdown), free-text Q&A written in the app, a shared Google Drive folder, and notes, documents or transcripts sent through the MCP update tool. Not available today: business email, CRM records, call-recording connectors, review sites, Notion, Confluence, Zendesk and Intercom.
Salespeak currently builds context from supported sources: your website (re-crawled daily), uploaded content, a shared Google Drive folder (synced on a schedule), and information your team sends through the MCP update tool. Not every company system is synchronized automatically today. We are expanding the sources and automation available over time.
Yes. The layer flags facts that disagree across your sources, and facts that have gone too long without review, so a person can rule on them. Salespeak also centralizes the context agents use, which makes the inconsistencies it cannot see, such as a claim in an asset that was never connected, easier to find and correct.
The GTM Context Layer returns source-backed records with authority and last-updated metadata, but those records are not automatic independent verification of a claim. It does not automatically update assets already written from older information, such as decks, pages or emails; a change does reach every later answer from the layer. It does not ingest business email, CRM records, call-recording systems, review sites, Notion, Confluence, Zendesk or Intercom today. Everything in the "How it works today" section is available now; the Company Context documentation lists the current tools.
Your team controls what goes in: which sources are connected, which files and Q&A entries are added, and which updates are sent. Updates through the MCP update tool come from a client signed in to your workspace. This page does not claim a separate approval step between an update and the agents that receive it; ask about it on the fit call.
MCP access uses OAuth 2.1 sign-in scoped to your workspace. The HubSpot app installs with no CRM scopes. Google Drive indexing covers only folders you share with the Salespeak service account, and the website crawl respects noindex. Salespeak's security program is on the security page. Ask on the fit call about retention and isolation for context-layer content.
A knowledge base stores information for people to find. A RAG system retrieves passages at the moment an agent asks. A GTM context layer is the maintained set of company facts behind those answers: which sources count, how changes get in, and which agents receive them. The approaches can work together; retrieval can sit underneath a context layer. Answers from the Company Context connector include the supporting records, such as source type and URL where available, authority label and last-updated time. Those records show provenance; they do not independently certify a fact as true. Changes are reflected in later answers from connected clients, but Salespeak does not automatically revise decks, website pages or emails that were already written.
You can build a first version by combining a knowledge base, retrieval, prompts, and agent connections. The harder part is maintaining it as your company changes: deciding which sources agents should rely on, finding disagreements and aging information, resolving changes once, and keeping multiple agents on the same version of the company. Salespeak gives you a dedicated layer for that work: one place to maintain approved company context, surface disagreements, and make the same context available to multiple agents. The goal isn't to replace every knowledge system you already have. It's to keep every AI agent from becoming its own separate version of your company.
Yes, it is generally available. There is no self-serve signup: it begins with a call. It starts at $2,000 per month on a month-to-month contract you can cancel anytime. The plans on the pricing page cover other Salespeak products, and are month-to-month too.
Then use it across:
Your agents will change. Your company truth shouldn't.
Create a shared, verified GTM Context Layer for the AI systems your company uses today, and the agents you will add tomorrow.
Where this stands: the GTM Context Layer is generally available. It has no self-serve signup, so it begins with a call. It starts at $2,000 per month, month-to-month. We have no customer case studies with numbers yet, and we won't invent them.
Read the operating guide to keeping GTM context up to date when company facts change.
Guides: company truth vs. account data, context layer vs. DIY RAG, and verified company knowledge for HubSpot Breeze agents.
It starts with a 30-minute call on which agents you run, which sources they need, and whether the GTM Context Layer fits.
