Agentic AI in retail: How Dynamics 365 powers Commerce Anywhere

Retail Frontier Firms are evolving their operating models to keep pace with increasingly dynamic markets, using AI to support more responsive and resilient decision-making and execution across commerce channels. Rather than improving individual functions in isolation, these organizations are rethinking how commerce operates end to end, enabling AI agents to work alongside people to support faster, more consistent outcomes across the business. This evolution is accelerating as retailers navigate rising customer expectations, sustained margin pressure, volatile demand, and ongoing labor constraints: conditions that benefit from decisions being made and executed more continuously.

In Retail Frontier Firms, AI capabilities are embedded where decisions and value are created: in stores, at the digital shelf, across merchandising, pricing, fulfillment, customer service, and checkout. AI agents interpret signals from customers, inventory, suppliers, and channels and help coordinate actions across the enterprise. This supports retailers as they respond to change with greater speed, consistency, and scale across touchpoints.

This operating model is enabled by agents that share context and operate cohesively across the retail ecosystem. At the core of agentic commerce is Model Context Protocol (MCP), which provides AI agents with access to a shared, enterprise-grade understanding of products, inventory, pricing, policies, and customer intent. By grounding agents in a common business context, MCP helps support aligned, governed, and consistent decision-making across channels and functions. The future of retail is increasingly shaped by a human and AI agent operating model, connected by shared context and open protocols.

• Model Context Protocol (MCP) unlocks hundreds of thousands of business functions for secure, real-time use by agents, developers, and applications. 
• Agent Communication Protocol (ACP) enables agents across merchandising, supply chain, store operations, and service to collaborate end to end, helping reduce fragmentation and align execution across functions.
• Payment and transaction agent protocols extend AI capabilities through checkout and settlement, supporting trusted, compliant transactions across in-store, online, and conversational commerce.

Together, these capabilities support a more outcome-driven operating model focused on availability, margin, conversion, service levels, and loyalty. Humans define strategy, priorities, and guardrails, while AI agents help orchestrate execution across day-to-day operations: supporting modern retail operations designed for Commerce Anywhere.

Emerging retail trends shaping the future of commerce

As consumer expectations continue to rise, shoppers increasingly demand seamless, continuous interactions where they move effortlessly from social-commerce discovery to mobile checkout, in-store pickup, curbside fulfillment, or voice-activated reordering. Frontier retail responds to this shift by dissolving the boundaries between channels and touchpoints, allowing commerce to adapt in real time to customer intent, location, and context. For brands, this means the ability to deliver frictionless, anticipatory commerce at scale by meeting customers wherever they are, with relevance and speed, without adding operational complexity.

The industry is rapidly shifting away from static, siloed channels toward autonomous, context-aware agents that orchestrate buying journeys seamlessly across stores, digital experiences, and conversational interfaces. Agents move beyond traditional personalization. They actively guide product discovery, shape contextual offers, negotiate availability, and coordinate fulfillment, helping to continuously optimize inventory, pricing, promotions, and supply-chain decisions behind the scenes. As personalization and automation become table stakes, agentic AI emerges as the strategic engine driving scalable growth and sustainable Commerce Anywhere.

Introducing Microsoft Dynamics 365 Commerce MCP Server

Agentic commerce introduces a new operating model in which AI agents collaborate through MCP, enabling continuous decision-making and coordinated execution across the retail value chain. The new Dynamics 365 Commerce MCP Server exposes core retail business logic including catalog, pricing, promotions, inventory, carts, orders, and fulfillment as MCP-enabled capabilities. Expected to be in preview in February 2026, this will allow retailers to build agentic commerce experiences where AI agents can securely discover, decide, and execute retail workflows across digital, physical, and conversational channels.

By combining the ERP, Analytics, and Commerce MCP servers, Dynamics 365 supports a more agent-driven operating model in which front-office experiences and back-office operations are connected and optimized, helping retailers operate with greater agility and readiness for Commerce Anywhere.

How retailers can begin adopting agentic commerce today

Retail leaders can begin moving toward agentic commerce by adopting AI agents in three practical ways:

1. Starting with agents embedded in Dynamics 365
2. Extending capabilities through custom-built agents using MCP
3. Leveraging partner-built agents across the broader retail ecosystem

Together, these approaches allow retailers to progress at their own pace while aligning agent adoption to their operating model, business priorities, and maturity.

1. Start with agents embedded in Dynamics 365

Purpose-built agents are designed to address common retail challenges and operational friction points. Dynamics 365 agents and retail industry agents can be embedded directly into core business processes, allowing teams to realize value quickly.

Microsoft retail industry agents, like the Catalog Enrichment Agent and Personalized Shopping Agent are examples of vertical-specific agents designed around retail data models, workflows, and decision patterns that support scenarios like product discovery, assortment accuracy, and personalized engagement without requiring custom development.

Today, in Dynamics 365 the Supplier Communications Agent is a good example of embedded agents in action. Retailers can proactively monitor supply signals and engage suppliers in real time to confirm availability, align delivery timelines, and respond to changes earlier. This supports faster coordination, fewer surprises, and more reliable execution at scale.

2. Build custom agents using MCP

Retail operations are shaped by business logic that is unique to each organization: driven by merchandising strategies, store formats, service models, and supply-chain constraints. Microsoft Copilot Studio enables retailers to build custom AI agents that encode their own rules across replenishment, allocation, fulfillment, and store execution, aligning agent behavior directly to how the business operates.

These custom agents can operate across planning and selling in the flow of work using MCP-powered access to enterprise systems. Inside Microsoft Teams, Merchandising Managers and Planners can collaborate in real time with agents that access products, demand forecasts, supplier relationships, inventory, and pricing through Dynamics 365 ERP MCP.

On the selling side, through the Dynamics 365 Commerce MCP Server, your custom agents can extend intelligence into customer experiences. Agents can discover products, personalize offers, assess availability, reserve inventory, and complete transactions across digital, physical, and conversational channels while operating with a unified view of pricing, promotions, and fulfillment.

3. Extend agentic commerce through partners and the ecosystem

Retailers can further accelerate agent adoption by leveraging partner-built agents designed for specific retail scenarios and industries. Commerce MCP enables software development companies and system integrators to build agents more quickly by reducing integration overhead, standardizing access to retail data, and maintaining enterprise-grade trust and compliance.

Early partner solutions already demonstrate the breadth of what’s possible, from store associate productivity and clienteling to conversational commerce and business-to-business (B2B) buying experiences, including:

• Amicis: The Store Commerce Agent is a voice-first, screen-aware assistant designed for in-the-moment store execution. It can help associates complete high-friction tasks like returns, exchanges, order lookups, and policy checks in Dynamics 365 Commerce using natural voice commands, while adapting to what’s on the POS screen.
• Evenica: The B2B Licensee Product Request Agent uses conversational AI and image recognition to support licensees in finding beverage products. When a product is not available in the catalog, the agent can create a request case to support the product intake process.
• Argano: The Retail Clienteling Agent offers a conversational clienteling experience by bringing together customer insights, product data, and agentic AI into a single, governed workflow. It helps retail associates improve customer relationships by delivering personalized, brand-aligned interactions before, during, and after in-store appointments.
• Sunrise: The Commerce Companion is a suite of retail agents that help simplify everyday store operations across inventory and fulfillment to purchasing and store processes. Using natural language, it is designed to deliver fast, accurate answers and guided actions, which can enable associates to serve customers efficiently while keeping operations moving smoothly.
• Visionet: FashionGPT Agent can turn natural-language shopping intent into real-time retail execution across product, pricing, inventory, and promotions. It drives the end-to-end shopping journey and help turn conversations into measurable actions across channels.

Together, embedded agents, custom-built agents, and partner solutions give retailers flexible entry points into agentic commerce supporting near-term impact while laying the foundation for a more adaptive, AI-enabled operating model across Commerce Anywhere.

Agentic retail with Dynamics 365 in action at NRF 2026

At NRF, we will demonstrate how Dynamics 365 works with Copilot and agentic capabilities to support Commerce Anywhere and more efficient, end-to-end retail operations. We will share examples of how retailers are using Dynamics 365 to evolve their operating models and advance Frontier Firm capabilities.

Visit us during NRF expo hours at Level 3, Booth 4503, and join the related theater sessions at our booth:

• Beyond the Boutique: How Frette Uses AI to Transform Store Experience
  January 11, 2026 (Sunday), 2:00 PM ET
  Session led by Sunrise Technologies
• Reimagine retail business processes with Agentic ERP
  January 13, 2026 (Tuesday), 2:30 PM ET

The future of retail belongs to frontier organizations that can sense, decide, and act in real time. With agentic commerce enabled by Dynamics 365, retailers gain the foundation to move faster with confidence, aligning strategy, execution, and customer experience through intelligent agents that operate seamlessly across every channel. We look forward to connecting with you in New York and exploring how agentic business applications in Dynamics 365 can support your next step forward.

The post Agentic AI in retail: How Dynamics 365 powers Commerce Anywhere appeared first on Microsoft Dynamics 365 Blog.