How to connect agents across your tech stack with MCP

The best GTM teams won’t be the ones with the most AI, but the ones with the best-connected AI.

These are the teams that know that it’s not enough to use disparate AI agents — real impact comes from shared context and interoperability. Without that, AI agents operate in silos, leaving revenue teams disconnected from the shared signals necessary for effective workflow transformation.

Previously, solving this problem required costly engineering work and fragile integrations. But not anymore.

That’s why we at Gong are excited to offer support for Model Context Protocol (MCP) (rolling out to customers in the next few weeks*). We’re launching MCP connectivity with industry leaders like Amazon, Anthropic, Google, HubSpot, Microsoft, OpenAI, Salesforce, and numerous other technology providers (browse them here) so you can connect Gong directly to the AI platforms and tools your team already uses.

Teams can configure Gong as both an MCP client (pulling external context into your revenue workflows in Gong) and an MCP server (enabling Gong's revenue-specific agents to work with every other AI agent in your stack).

Configuring Gong as an MCP client means Gong Agents, like AI Briefer, can bring external data directly into the revenue workflows your teams already use. Support ticket info, intent signals, buyer engagement, product usage, business metrics, and more can now show up inside of Gong AI Briefs, giving reps and customer teams the context they need to take the right next step.

Here are a few ways you can bring outside context into Gong to better streamline your revenue workflows:

1. High-fidelity handovers: You can enrich AI Briefer with real-time support ticket data. Instead of walking into a renewal call without the right info, AEs get a brief beforehand that highlights unresolved technical issues alongside the latest conversation themes.
2. Full-spectrum account health: By pulling in external intent signals, Gong can help teams spot account health changes and deal risk earlier, with external signals layered alongside revenue context.
3. Deep business intelligence: By connecting BI tools directly to AI Briefer, you bring quantitative KPIs (e.g., renewal numbers, financial data) into the same view as your qualitative conversation insights.
4. Real-time product adoption: You can connect product usage tracking tools to Gong to augment your briefs with adoption metrics. This allows CSMs to pinpoint exactly where a customer sees value before meetings kick off.

As an MCP server, Gong lets other platforms call on Gong-generated outputs like account summaries, deal risks, objections, and next steps. So whether your team is working in Claude, Microsoft Dynamics 365, ChatGPT, Agentforce, Amazon Quick, or another external AI system, Gong can hand off revenue context that has already been analyzed and is ready to use. While the end of manual copy-pasting is an obvious win, the bigger win lies beneath.

Without Gong, external AI systems and agents often have to start from scratch — interpreting transcripts, piecing together CRM data, and relying on whatever context was manually shared with them. Gong changes that by giving external agents a trusted handoff from the revenue workflows already happening in Gong.

Because Gong delivers authoritative answers grounded in expert revenue context, external agents can build on trusted revenue intelligence instead of interpreting fragmented inputs or having to reconstruct the story.

Gong as an MCP Server can power connected workflows across many systems. Here are two of the most common places teams can bring Gong context into:

The primary benefit here is seamless execution. You can query Gong about accounts and deals directly from your CRM to remain action-oriented in your agentic workflows. A few examples of what you can do:

• Have a Dynamics 365 lead qualification agent use Gong’s account context to understand what business pain was discussed in previous customer conversations, then route the lead with the context attached.
• Integrate Agentforce into sales workflows to surface comprehensive deal briefings and instant customer insights—keeping reps equipped with critical context and focused on their highest-priority accounts.
• Ask a HubSpot Breeze Agent to use Gong to name the decision maker on a deal. Then have it identify that person’s level of engagement, objections they mentioned, and next steps that would address those objections.

Connecting your LLMs and AI assistants to Gong does not turn them into revenue experts on their own. Rather, it gives them a continuous handoff from the Gong Agents already working on your revenue data, so they can help with planning, research, and content creation using context that is already structured for revenue teams. Here’s how:

• Use a platform like Claude to retrieve account or deal summaries across a set of target accounts from Gong, helping teams quickly compare stakeholders, objections, risks, and next steps.
• Use Microsoft Copilot Studio to bring deal risks and conversation themes that have been identified by Gong into an internal workflow, helping teams understand which opportunities need attention and why.
• Make ChatGPT a strategic partner when you draft account plans or QBR decks with Gong-generated briefs and customer context. By accessing Gong via MCP, ChatGPT’s outputs reflect recent account activity, customer priorities, pain points and next steps.
• Use Amazon Quick to combine Gong insights with other enterprise knowledge sources through MCP — for example, showing how implementing a product feature request could impact pipeline opportunities.

Gong as an MCP Server can also extend revenue context into the tools teams use for enablement, onboarding, support, and customer engagement. For example, an enablement platform could use Gong-generated deal context to recommend the right content for a competitive objection, or a customer workflow that could bring recent account risks into the next best action for a CSM.

By launching MCP connectivity with industry leaders like Microsoft, Salesforce, HubSpot, Google, OpenAI, and Anthropic, we’re ensuring that Gong isn’t just a part of your tech stack. It’s the connective layer that helps agents across your stack act on Gong’s expert revenue context, powering more efficient workflows and delivering real value from your agentic investments.

Ready to see what a connected GTM stack looks like? Take a demo or reach out to your account team to see how Gong’s support for MCP can transform your revenue motions.

*Gong MCP Server is GA to all Gong customers with Gong Credits. Credits are rolling out in phases throughout the month of June.