by Richard Sonnen, Sentinel’s Director of AI Innovation and Consulting

Despite massive investments in AI technology, many organizations have struggled to realize its promised business value. A critical barrier stands in the way: the inability of AI systems to effectively interact with the business tools and data sources where valuable information lives. Without a standardized approach, companies resort to building one-off, fragmented integrations that are expensive to maintain and impossible to scale.

The Model Context Protocol (MCP), introduced by Anthropic in November 2024, directly addresses this challenge by providing an open standard for AI-to-tool interactions. Much like USB-C revolutionized connectivity in hardware, MCP creates a universal interface that allows any AI model to communicate with, take actions through, and retrieve context from virtually any business system.

For IT leaders focused on practical AI implementation, MCP solves one of the most challenging technical hurdles in enterprise AI deployment by:

• Standardizing how AI uses business tools to retrieve data and accomplish tasks
• Simplifying the integration landscape by reducing the number of custom connections needed
• Creating reusable connections that work across different AI models and workflows
• Maintaining context as AI moves between different business applications

The rapid adoption of MCP by industry leaders including OpenAI, Microsoft, Google and numerous enterprise software providers signals its emergence as a foundational standard for AI implementation strategies in 2025 and beyond.

The Business Case for MCP

As organizations consider whether to invest in MCP implementation, the business justification comes from three critical aspects: connecting to valuable data, enabling direct tool interaction, and standardizing integration across systems.

Breaking Down Information Silos

A significant limitation facing enterprise AI today is the inability to access and utilize the organization's most valuable information assets. Your competitive advantage doesn't come from using the same AI models as everyone else; it comes from applying those models to your unique business data.

MCP breaks down these information silos by enabling AI to:

• Access real-time data from your CRM, ERP, and knowledge management systems
• Maintain context while navigating between different enterprise systems
• Deliver insights based on your proprietary business information
• Transform from generic capabilities to organization-specific expertise

Enabling Active Tool Use

MCP goes beyond passive data access, allowing AI systems to actively use business tools and take actions within your organization's systems. Unlike approaches that simply retrieve information, MCP creates a bidirectional channel for AI to execute operations across your business applications.

This tool-centric capability transforms AI from an isolated question-answering system into an active participant that can:

• Execute tasks in your CRM, ERP, and custom applications
• Update records, create tickets, and manage workflows
• Analyze data across systems and synthesize actionable insights
• Coordinate sequences of operations across multiple business tools

Measurable ROI Through Standardization

Before MCP, organizations needed custom integration work for each combination of model and data source—what engineers call an "M×N problem." This approach created fragmented implementations, duplicated efforts, and significant maintenance burdens.

MCP transforms this into an "M+N" solution where teams can reuse standardized connections across systems, delivering measurable ROI through:

• Significant reduction in integration development time
• Lower maintenance costs for AI connections
• Faster deployment of AI capabilities across the organization
• More flexible architecture that adapts as needs evolve

Real-World Business Impact

The true test of any technology standard is how it performs in production environments. Organizations across diverse industries have already begun implementing MCP and seeing tangible benefits from standardized AI-to-tool interactions.

Block (Square): Enhancing Merchant Experiences

Block (formerly Square) has implemented an official Square MCP server providing secure access to their Connect API. This implementation allows AI systems to interact directly with transaction data, customer information, and financial records, but also to actively perform operations across Square's extensive business ecosystem.

The MCP server enables AI assistants to execute actions including:

• Processing payments and managing checkout operations
• Creating and updating customer records
• Managing catalog items, categories, and inventory
• Scheduling appointments and handling bookings
• Generating and processing invoices
• Tracking order fulfillment

These capabilities have transformed how merchants operate. Block's CTO Dhanji R. Prasanna has endorsed this approach, describing open technologies like MCP as "bridges that connect AI to real-world applications, ensuring innovation is accessible, transparent, and rooted in collaboration." He specifically highlighted how MCP enables "agentic systems, which remove the burden of the mechanical so people can focus on the creative."

Atlassian: Transforming Knowledge Work

Atlassian's implementation of a remote MCP server for Jira and Confluence Cloud demonstrates how the protocol transforms knowledge work environments. Beyond simply accessing data, their MCP implementation enables AI systems to actively interact with and modify content across their collaboration tools.

Through their MCP server, AI systems can:

• Create new Confluence pages with synthesized information
• Generate and populate Jira work items in bulk
• Produce summaries of work distributed across Atlassian products
• Update documentation based on project progress
• Surface relevant knowledge in context of specific tasks

CTO Rajeev Rajan emphasized that MCP aligns perfectly with Atlassian's strategy of being "open by design," noting that "MCP hits every one of these notes" in their approach to building healthy software ecosystems. This implementation reduces context switching, accelerates decision-making, and ultimately allows teams to spend more time on meaningful work rather than administrative tasks.

Fitting MCP Into Your AI Strategy

Successfully implementing MCP requires more than just technical understanding - it demands strategic thinking about how this capability fits into your broader AI and digital transformation initiatives. The most successful adopters view MCP not as a standalone technology but as a critical component in their AI architecture.

Aligning With Digital Transformation

For organizations pursuing digital transformation, MCP serves as a crucial architectural component that bridges AI capabilities with operational business systems:

• Enhances contextual intelligence: By connecting models to proprietary information, AI transitions from generic capabilities to understanding your specific business context and data.
• Enables process automation: MCP allows AI to execute tasks within business systems, automating routine processes while maintaining awareness of your organization's unique workflows.
• Creates adaptable architecture: As your digital ecosystem evolves with new data sources and tools, MCP provides a standardized way to connect them to AI without extensive reconfiguration.
• Orchestrates cross-system interactions: The protocol enables AI to coordinate both information retrieval and actions across multiple business applications, breaking down traditional system boundaries.

Adoption Roadmap for IT Leaders

The path to MCP implementation should be methodical and in alignment with your organization's broader AI strategy. Rather than viewing it as a massive undertaking, successful organizations approach MCP adoption as a progressive journey with measurable milestones.

Where to Begin Your MCP Journey

Rather than attempting enterprise-wide implementation immediately, start with focused, high-impact projects that demonstrate MCP's value:

1. Automate routine workflows: Identify repetitive, rule-based processes in departments like customer service or finance where AI could execute actions across systems using MCP connections.
2. Enhance existing AI applications: If you've already deployed AI for analytics or insights, add MCP to enable those systems to initiate actions based on their findings.
3. Connect to pre-built MCP servers: Major platforms offer ready-made MCP servers that can be quickly integrated with minimal development.
4. Start with a narrow scope: Choose a single business process with clear ROI potential rather than trying to connect all your systems at once.

Successful MCP implementations begin with focused projects that demonstrate value through measurable outcomes like time saved or error reduction, then expand methodically across the organization.

Leveraging the Growing Ecosystem

The rapidly expanding MCP ecosystem significantly accelerates implementation and reduces risk. Rather than building every integration from scratch, you can leverage ready-made components that have been battle-tested across multiple deployments:

• Vertical-specific integrations: Industry-focused MCP servers for healthcare, finance, manufacturing, and retail systems that connect to specialized tools and data sources
• Data connectors: Ready-made servers for databases (SQL, NoSQL), data warehouses, and analytics platforms
• Productivity tools: Integrations with document management, email, calendar, project management, and team collaboration systems
• Development environments: MCP servers for code repositories, CI/CD pipelines, and deployment systems
• Enterprise platforms: Connections to ERP, CRM, HRIS, and supply chain management systems
• Custom server frameworks: SDKs in Python, TypeScript, C#, Java, and other languages for building tailored MCP servers

This pre-built ecosystem reduces your time-to-value from months to weeks, minimizes development costs, and allows your team to focus on business-specific customizations rather than core connectivity.

Positioning for the Future

As AI capabilities continue to evolve, your organization's ability to extract business value will increasingly depend on how effectively you integrate these systems into your operations. MCP provides a strategic foundation that helps future-proof your AI investments while accelerating time-to-value.

Strategic Considerations for 2025-2026

As you develop your AI roadmap for the next 18 to 24 months, consider these immediate actions:

1. Assessment: Identify high-value connection points between AI systems and business data sources
2. Capability building: Ensure your teams understand MCP's potential and practical implementation approaches
3. Partner selection: Engage with partners like Sentinel, who bring both technical expertise and strategic perspective
4. Pilot planning: Design initial implementations focused on specific business problems that demonstrate value

By standardizing how AI systems connect to and interact with organizational systems, MCP offers a path to more capable AI that becomes an active participant in your business processes. For IT leaders focused on extracting real business value from AI investments, MCP represents not just a technical standard but a strategic foundation for transforming AI from passive analysis to active business operations.

Sentinel stands ready to help your organization with all of its AI needs, including the development of an AI strategy and the proper utilization of MCP standards in your IT environment. Contact us or your Sentinel Account Manager today to learn more!