Observed Signal · Mar 9, 2026 · Technical Release · Source: AdExchanger · Impact: 4/5 · Sentiment: Positive
MCP: The Game-Changer for AI in Advertising
The Model Context Protocol (MCP), an open-source specification developed by Anthropic and released in late 2024, standardizes how large language models (LLMs) connect to external tools and software. MCP defines which tools a model may use, how to call them, and what data each action touches, enabling AI agents to chain actions (e.g., pull reports, create audiences, launch campaigns) across multiple systems. Industry-specific protocols such as AdCP (advertising), WebCP (web actions) and UCP (commerce) are being built on top of MCP. Early ad tech adopters include Criteo, Similarweb, Adverity and MadConnect, which use MCP to enable cross-platform workflows, conversational state and orchestration. While MCP promises reduced integration complexity and better memory/orchestration for campaign workflows, challenges remain around governance, business incentives, data quality and widespread implementation.
MCP is an open standard for how LLMs connect to external tools; widespread adoption could enable cross-platform AI orchestration in ad tech, reduce bespoke integrations, and change campaign planning/activation workflows, though governance and data-quality barriers remain.
Track Anthropic Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your free Explorer workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- Anthropic developed and open-sourced the Model Context Protocol (MCP) in late 2024.
- MCP is a specification that standardizes how LLMs interact with external tools, defining allowed tools, call formats, and data touched by actions.
- Industry-specific protocols—AdCP (advertising), WebCP (web actions) and UCP (commerce)—use MCP as an underlying transport.
- Early ad tech implementations include Criteo, Similarweb, Adverity and MadConnect using MCP to enable cross-platform AI workflows and preserved conversational state.
- MCP has been downloaded tens of millions of times, but adoption in ad tech remains nascent and faces governance and data-quality challenges.
Connected Companies & Entities
6 Entities mappedOntology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
Adtech Firms Connect MCP Servers to Chat Assistants
Adtech vendors are beginning to expose advertising data and functions to large language model (LLM) workflows using the Model Context Protocol (MCP). German firm Welect and Canadian company Stackadapt have each launched MCP servers: Welect focuses on exposing domain-specific knowledge (including signal-loss info, inventory sources and its Choice‑Driven Advertising approach) to chat assistants, while Stackadapt makes campaign-level data (performance, pacing, creatives, insights) available for prompt-driven querying and analysis. A separate integration between PubMatic and Adroll demonstrates cross-platform troubleshooting via MCP, enabling diagnostic data to flow between DSP/SSP systems and LLMs. The article frames MCP as an early-stage, fragmented but promising bridge that shifts interfaces from vendor dashboards toward conversational workflows; current use cases emphasize data access and decision support rather than full automation of bookings or programmatic buys.
Anthropic's Model Context Protocol (MCP) Explained
Model Context Protocol (MCP) is an open standard introduced by Anthropic that standardizes how applications provide external tool context to large language models (LLMs). MCP defines three core components — host, client, and server — and lets tool providers implement MCP-compatible servers so any MCP-speaking host can access tools without custom integration code. The protocol reduces the maintenance burden that arises when many different tools and provider APIs must be integrated, because updates are handled by the tool provider's MCP server rather than each host. The article outlines the MCP request/response flow and gives a minimal example (a weather MCP server exposing get_alerts and get_forecast) to demonstrate how hosts like Cursor can call MCP servers via an MCP client and return grounded results to an LLM.
zMaticoo Launches Model Context Protocol
zMaticoo announced the Model Context Protocol (MCP), a new API protocol designed to give large language models (LLMs) tool-oriented access to business advertising data. Built on an Open API and optimized for AI scenarios, MCP allows AI agents to read, write and operate ADX/DSP data via natural-language commands. The release highlights secure, token-based core tools (adx-report and dsp-report), a simplified three-step integration plus one-click Agent access, and immediate availability for integration. zMaticoo says customers should contact their business representative to obtain tokens and begin using MCP for AI-driven ad data operations.
Track Real-Time Market Signals & Shifts
Set up custom watchlists to receive automated, evidence-grounded executive digests whenever material signals or shifts occur across your tracked landscape.