Unlocking NetSuite's AI potential: The Model Context Protocol revolution

Article    August 18, 2025
SHARE
SHARE
BOTTOM LINE UPFRONT

NetSuite’s new Model Context Protocol (MCP) is an AI breakthrough that enables LLM agents to directly access and analyze NetSuite business data in real time—addressing month-end close delays, cash flow visibility gaps, and demand planning disconnects. Learn how to leverage NetSuite’s MCP server with data privacy top-of-mind, including a step-by-step guide for how to integrate the server with Claude.

There’s been a breakthrough in enterprise AI integration. NetSuite’s new Model Context Protocol (MCP) server represents the missing link that businesses have been waiting for: secure, direct AI access to their critical business data. For the first time, organizations can enable AI systems to intelligently query and analyze their NetSuite data in real time, opening unprecedented opportunities for automation, insights, and operational efficiency.

In other words, MCP technology transforms NetSuite from a data repository into an intelligent business partner—while maintaining the security and control that CFOs need. This article breaks it down:

The Challenge: Critical business pain points

CFOs face mounting pressure to accelerate financial processes while maintaining accuracy and control. Three challenges consistently emerge as top priorities:

  • Month-end close bottlenecks: Finance teams spend the first week of each month manually gathering data from multiple NetSuite modules, validating transactions, and reconciling accounts. What should be a streamlined process becomes a marathon of spreadsheet exports, manual calculations, and email chains seeking missing information.
  • Cash flow blind spots: Without real-time visibility into payment timing, invoice aging, and working capital positions, CFOs make decisions based on outdated snapshots. By the time traditional reports surface cash flow issues, opportunities for optimization have already passed.
  • Demand planning disconnects: Sales forecasts, inventory levels, and financial projections exist in separate silos. Finance teams struggle to connect operational metrics with financial impacts, leading to reactive rather than predictive planning.

Despite significant investments in AI technology, most organizations can’t leverage available tools to address these fundamental challenges. The primary obstacle? Data accessibility. Critical business information remains locked in NetSuite, forcing employees to manually extract, format, and interpret data before AI can provide value.

Traditional approaches to AI-ERP integration require:

  • Complex custom development projects
  • Significant IT resources and ongoing maintenance
  • Security compromises or administrative access risks
  • Limited query capabilities and static data exports

The result: AI remains disconnected from the real-time operational data that could solve these pressing business problems.

The Solution: Model Context Protocol

NetSuite’s MCP is the missing foundation. The server doesn’t solve these business challenges directly—instead, it provides the critical groundwork that enables AI solutions to address them. Think of MCP as the essential bridge that finally allows AI systems to “see” and interact with your NetSuite data in meaningful ways.

MCP serves as the intelligent interface that:

  • Enables structured queries: AI can ask sophisticated questions about your business data and receive precise, contextual answers.
  • Provides real-time context: AI systems gain access to current business state, not just historical snapshots.
  • Maintains data integrity: Information flows through secure, controlled channels designed for AI consumption.
  • Creates scalable access: Multiple AI applications can leverage the same secure data interface.

With MCP as the foundation, organizations can now build sophisticated AI solutions that were previously complex, expensive, and time-consuming to implement—solutions that directly address month-end close delays, cash flow visibility gaps, and demand planning disconnects.

MCP’s transformative business benefits

Intelligent context assembly

MCP transforms how AI systems understand your business by providing rich, contextual data rather than isolated data points. Instead of AI making decisions based on limited snapshots, MCP enables comprehensive context gathering that connects transactions, relationships, and patterns across your entire NetSuite environment. This contextual intelligence allows AI to provide insights that reflect the true complexity of your business operations.

Optimized data exchange

Traditional data access methods force AI systems to work with whatever static information they can access. MCP enables dynamic, intelligent queries that retrieve precisely the information needed for each analysis. This optimization reduces processing overhead, improves response accuracy, and ensures AI systems receive data in formats that maximize analytical capability.

Scalable intelligence infrastructure

MCP creates a standardized foundation for AI applications to access NetSuite data efficiently. Rather than building custom integrations for each AI solution, organizations can leverage MCP’s optimized data exchange protocols to rapidly deploy and scale multiple AI applications. This infrastructure approach transforms AI implementation from a series of complex custom projects into a streamlined capability expansion.

Real-world applications enabled by MCP

Enhanced decision-making capabilities

With MCP providing optimized context gathering, AI applications can access comprehensive business intelligence that spans multiple NetSuite modules simultaneously. This enables more nuanced analysis and recommendations that account for the full scope of business operations rather than isolated data points.

Process automation opportunities

MCP’s efficient data exchange protocols enable AI systems to continuously monitor business processes and provide real-time insights. Organizations can deploy AI applications that understand operational context and deliver actionable intelligence precisely when decisions need to be made.

Strategic planning enhancement

The rich contextual data available through MCP allows AI systems to connect operational metrics with financial outcomes, enabling more sophisticated forecasting and scenario planning. Finance teams gain access to insights that bridge departmental silos and reflect true business complexity.

Note: MCP doesn’t solve all of these challenges itself. Rather, it provides the essential data access foundation that makes sophisticated AI solutions accessible and cost-effective for the first time.

Implementation strategy: Enterprise-grade security

The most common CFO concern about AI-ERP integration centers around data security and compliance risk. Direct connections between AI systems and production environments create legitimate concerns about data exposure, access control, and regulatory compliance.

Smart implementation strategies address these concerns through layered security architectures. Rather than connecting AI directly to production NetSuite environments, leading organizations implement enterprise-grade intermediary solutions that provide:

  • Data obfuscation: Sensitive information is masked or tokenized before AI analysis.
  • Access sandboxing: AI systems operate in controlled environments with precisely defined data boundaries.
  • Enterprise infrastructure: Dedicated cloud infrastructure ensures data never leaves your controlled environment.
  • Audit transparency: Complete visibility into what data AI accesses and how it’s used.
  • Compliance automation: Built-in controls for regulatory requirements across multiple jurisdictions.

The key is choosing implementation partners who understand that security isn’t just about technical controls—it’s about creating architectures that give CFOs confidence while enabling AI transformation. Accordion’s approach combines enterprise-grade security protocols with advanced MCP optimization to deliver both confidence and performance.

Implementation best practices include:

  1. Creating dedicated AI roles: Establish custom NetSuite roles with precisely limited permissions for AI access.
  2. Implementing query optimization: Leverage advanced tooling to ensure efficient data access and cost control.
  3. Deploy secure intermediary solutions: Use enterprise-grade platforms that provide security layers between AI and production data.
  4. Maintain comprehensive audit trails: Track all AI queries and data access for compliance and security monitoring.
  5. Scale progressively: Begin with specific use cases and expand capabilities based on proven value and security confidence.

The competitive advantage: Advanced optimization capabilities

While MCP provides powerful baseline capabilities, organizations that invest in advanced optimization can achieve significantly greater efficiency and cost-effectiveness. NetSuite’s new “Custom Tool Scripts” SuiteScript type enables skilled developers to extend MCP server functionality beyond standard queries, creating tailored solutions that streamline complex business processes.

At Accordion, our approach leverages these Custom Tool Scripts to build proprietary optimization methods that:

  • Reduce query complexity for common business functions
  • Improve response times through intelligent data preprocessing
  • Minimize usage costs through efficient data access patterns
  • Create custom business logic that understands your specific operational context

This advanced customization capability represents a substantial competitive advantage in AI implementation speed and ROI. Organizations working with partners who understand both MCP’s technical architecture and Custom Tool Scripts development can achieve optimization levels that aren’t available through standard implementations.

Looking Forward: The future of intelligent ERP

MCP represents just the beginning of AI-ERP integration. As this technology matures, we anticipate:

Expanded Oracle ecosystem integration: Connecting AI across the full range of Oracle business applications.
Predictive analytics: Moving from reactive reporting to proactive business guidance.
Automated workflow optimization: AI-driven process improvements based on real-time operational data.
Enhanced decision-making: Executive dashboards powered by continuous AI analysis.

The time to act is now

Model Context Protocol technology represents a paradigm shift in how businesses can leverage AI with their operational data. Organizations that move quickly to implement MCP-enabled AI solutions will gain significant competitive advantages in operational efficiency, decision-making speed, and cost management.

The question isn’t whether AI will transform how businesses interact with their ERP systems. It’s whether your organization will lead or follow in this transformation.

Need support integrating NetSuite MCP with your LLM? Accordion laid out a guide that walks you through every step of your MCP integration with Claude AI, with data security top-of-mind:

The step-by-step NetSuite MCP setup guide for Claude AI

This guide walks you through setting up the NetSuite Model Context Protocol (MCP) integration with Claude AI, enabling you to interact with your NetSuite data using natural language conversations and perform tasks such as creating records, running queries, and accessing NetSuite data directly through Claude AI.

Prerequisites

Before starting the setup process, ensure you have:

NetSuite requirements

  • NetSuite account: Administrative access to your NetSuite environment
  • Integration features enabled: NetSuite requires specific features to be enabled for OAuth and SuiteTalk functionality. You must enable REST Web Services and OAuth 2.0 authentication

To enable these features:

    1. In NetSuite, go to Setup → Company → Enable Features
    2. Click the SuiteCloud subtab
    3. In the SuiteTalk (Web Services) section, check REST Web Services
    4. In the Manage Authentication section, check OAuth 2.0
    5. Click Save

Claude AI requirements

  • Claude AI Pro subscription: Required to use custom connectors and MCP functionality
  • Why Claude AI: Claude AI is the superior choice for this integration because:
    • No Usage Billing: Unlike other LLMs, Claude doesn’t charge based on token usage for MCP interactions
    • Strict Data Privacy: Anthropic has stringent data privacy practices with 30-day default retention and explicit user consent requirements for data training
    • Optimal MCP Support: Built specifically to work seamlessly with Model Context Protocol

Part 1: Installing MCP tools in NetSuite

Step 1: Access SuiteApp Marketplace

  1. Log in to your NetSuite account
  2. Navigate to Customization → SuiteCloud Development → SuiteApp Marketplace
  3. In the Search SuiteApps field, enter: MCP Tools
  4. Click the MCP Tools icon in the search results

Step 2: Install MCP Tools

  1. On the SuiteApp details page, click Install (located in the top right)
  2. Follow the installation prompts
  3. Wait for the installation to complete

** Important: If the installation fails, the MCP client will show an empty list of tools when you try to connect from Claude AI.

Part 2: Creating a custom role for MCP integration

Why a custom role is required

Critical security note: Administrators are not allowed to work directly with MCP for security reasons. A custom role ensures:

  • Proper permission management and security
  • Controlled access to NetSuite features through the AI integration
  • Compliance with NetSuite’s security framework

Step 1: Create the custom role

  1. In NetSuite, go to Setup → Users/Roles → Manage Roles
  2. Click New to create a new role
  3. Give your role a descriptive name (e.g., “MCP AI Integration User”)

Step 2: Assign Required Setup Permissions

Navigate to the Setup subtab of the Role record and assign Full level permissions for:

Minimum Required Permissions:

  • Log in using OAuth 2.0 Access Tokens
  • MCP Server Connection
  • OAuth 2.0 Authorized Applications Management

These three permissions are the absolute minimum required to establish the MCP connection.

Step 3: Assign functional permissions

Based on your intended use cases, add permissions for the data and records you want to access through Claude AI:

Common permissions to consider:

  • Transactions: For creating sales orders, invoices, payments, etc.
  • Lists: For managing customer, vendor, item, and other list records
  • Reports: For running and accessing reports
  • Custom Records: If you plan to work with custom record types
  • SuiteQL: For advanced querying capabilities

Permission strategy: The MCP server can run SuiteQL queries out of the box, but you must provide record-level permissions for the integration to query specific record types. Without proper permissions, functionality will be severely limited.

Step 4: Assign Role to Users

  1. Save the custom role
  2. Assign this role to users who will work with the MCP integration
  3. Ensure users have this role active when using the integration

Part 3: Connecting Claude AI to NetSuite

Step 1: Access Claude AI Settings

  1. Go to https://claude.ai/
  2. Ensure you have a Pro subscription active
  3. Click on your Profile (usually in the top right)
  4. Select Settings from the dropdown menu

Step 2: Add custom connector

  1. In Settings, look for Add Custom Connector option
  2. Click to add a new connector

Step 3: Configure connection details

Name: Enter a descriptive name for your connector

  • Example: “NetSuite Production” or “NetSuite – [Company Name]”
  • You can include an identifier to distinguish multiple NetSuite environments

URL: Enter your NetSuite MCP endpoint URL using this format:

https://<account_id>.suitetalk.api.netsuite.com/services/mcp/v1/all

Finding your Account ID: Your account ID is typically visible in your NetSuite URL when logged in, or can be found in Setup → Company → Company Information.

Step 4: Establish connection

  1. Click Connect
  2. You’ll be redirected to NetSuite for OAuth authentication
  3. Log in with credentials that have the custom MCP role assigned
  4. Adjust role: If you have multiple roles, ensure you select the custom MCP role you created
  5. Click Allow to authorize the connection
  6. You should see a success response confirming the connection

Step 5: Verify Integration

  1. An automatic integration record will be created in NetSuite
  2. To view it, go to Setup → Integrations → Manage Integrations
  3. You should see a new integration record for your Claude AI connection

Part 4: Using the NetSuite MCP Integration

Available Tools

Once connected, you’ll have access to various tools based on your role permissions. The available tools include:

  • Record management: Create, read, update, and delete records
  • SuiteQL queries: Advanced querying of NetSuite data
  • Transaction processing: Handle sales orders, invoices, payments, etc.
  • Reporting: Access and generate reports

For a complete list of predefined tools, see NetSuite MCP tools documentation.

Sample Conversational Prompts

Here are examples of how to interact with NetSuite through Claude AI:

Creating records:

  • “Create a new customer record for ABC Company with email contact@abccompany.com
  • “Add a new inventory item called ‘Widget Pro’ with SKU WGT-001”

Querying data:

  • “Show me all open sales orders from this month”
  • “What are our top 10 customers by total sales?”
  • “Find all items with inventory below 50 units”

Reporting:

  • “Generate a sales summary report for Q4”
  • “Show me aged receivables for customers over 60 days”

Best practices

  1. Be specific: Provide clear, detailed instructions for better results
  2. Use natural language: The AI understands conversational requests
  3. Verify results: Always review the output, especially for critical business operations
  4. Test permissions: Start with read-only operations to ensure everything works correctly

Troubleshooting NetSuite MCP

Empty tools list

Problem: After connecting, Claude AI shows no available tools.

Root cause: This is almost always a permission issue.

Solutions:

  • Verify the MCP Tools SuiteApp installed successfully
  • Check that your user has the proper custom role assigned and active
  • Ensure the role has the three required Setup permissions (OAuth 2.0 Access Tokens, MCP Server Connection, OAuth 2.0 Authorized Applications Management)
  • Confirm you’re using the correct account ID in the URL

Limited functionality

Problem: Connected successfully but can only perform basic queries, not access specific records.

Root cause: Insufficient record-level permissions in the custom role.

Solutions:

  • Review and expand the custom role permissions
  • Add specific transaction, list, or record permissions as needed
  • Remember: The role must include permissions for any records/features you want to access through Claude

Authentication failures

Problem: Unable to authenticate or connection fails.

Solutions:

  • Verify your NetSuite account credentials
  • Ensure your user account has the custom MCP role assigned and active
  • Check that OAuth 2.0 permissions are set to “Full” in the role
  • Confirm the account ID in the URL is correct
  • Verify that OAuth 2.0 is enabled in your NetSuite instance

Integration not showing in NetSuite

Problem: No integration record appears in “Manage Integrations”.

Solutions:

  • Refresh the Manage Integrations page
  • Verify the connection was completed successfully in Claude AI
  • Check that OAuth 2.0 Authorized Applications Management permission is enabled

Security considerations

Data privacy and LLM comparison

When connecting your NetSuite environment to an LLM, security and privacy are paramount. Here’s how the top 3 LLMs compare:

Claude AI (Anthropic) – Recommended

  • Data training: By default, user data is not used for model training. Users must explicitly opt-in
  • Data retention: 30-day default retention period, with user control over conversation deletion
  • Privacy controls: Limited employee access with strict controls and encryption at rest and in transit
  • Billing: No usage-based charges for MCP interactions

OpenAI ChatGPT

  • Data training: Enterprise customers can opt out, but consumer data may be used for training unless explicitly opted out
  • Data retention: Enterprise: Admin-controlled retention; Consumer: Varies by service type
  • Privacy controls: Enterprise customers get stronger protections than consumer users
  • Billing: Token-based usage charges can become expensive

Google Gemini

  • Data training: Consumer data used for training by default, with human reviewers analyzing conversations
  • Data retention: 18-month default retention, customizable to 3 or 36 months
  • Privacy controls: Integrated with broader Google data ecosystem; complex privacy settings
  • Billing: Usage-based pricing for enterprise features

Key security features

  • Role-based access: The integration respects NetSuite’s role-based permission system
  • OAuth 2.0: Secure authentication without sharing passwords
  • Audit trail: All actions performed through the integration are logged in NetSuite
  • Permission boundaries: The AI can only access data and perform actions allowed by the assigned role
  • Encryption: All data transmission is encrypted using industry standards

Important security implications

** Critical considerations:

  1. Data exposure: Any data accessible through your NetSuite role can potentially be sent to the LLM
  2. Business logic: Sensitive business rules and calculations may be revealed through queries
  3. Competitive information: Sales data, pricing, and customer information will be accessible
  4. Compliance: Ensure your organization’s data governance policies allow external AI processing

Recommended safeguards:

  • Use the most restrictive role permissions possible
  • Start with read-only access and expand gradually
  • Never include sensitive information in prompts
  • Regularly audit integration usage through NetSuite logs
  • Consider using Claude AI due to its superior privacy protections

Getting Help

  • NetSuite MCP documentation: Custom tool creation guide
  • Claude AI support: Available through your Claude AI account
  • NetSuite support: For SuiteApp installation and role configuration issues

Next Steps

Once your integration is set up and working:

  1. Explore capabilities: Try different types of queries and operations
  2. Create custom tools: Consider developing custom tools for your specific business needs
  3. Train your team: Share this guide with team members who will use the integration
  4. Monitor usage: Keep track of how the integration is being used and optimize permissions accordingly
  5. Regular security reviews: Periodically audit the role permissions and integration usage

Privacy and security summary: This integration provides powerful natural language access to your NetSuite data. Claude AI offers the strongest privacy protections among major LLMs, with no default data training, strict access controls, and user-controlled retention periods. Still, always consider the sensitivity of your data and implement appropriate safeguards through careful role permission management.

Ready to explore how MCP can transform your NetSuite environment? Accordion specializes in implementing enterprise-grade MCP solutions with advanced Custom Tool Scripts optimization. Get in touch to discuss implementation strategies tailored to your organization’s specific needs and value creation goals.

Need help with your NetSuite MCP build-out? Let's talk.

Our contact form is currently blocked by your cookie preferences. Please change your preferences to continue.