D365 F&O ↔ Akeneo PIM
Integration Architecture
Bridging ERP transactional data with product enrichment across 70+ operating companies, 4 business groups, and 17+ D365 legal entities at The Heico Companies
Integration Architecture Overview
End-to-end data flow from ERP source of truth through enrichment hub to multi-channel distribution, with a hot data bypass path for real-time pricing and inventory.
D365 F&O
- SKU / Item Number
- Pricing & List Price
- Inventory / ATP
- Cost & Vendor
Middleware
- Field Mapping
- Delta Detection
- Error Handling
- Queue Management
Akeneo PIM
- Descriptions & Specs
- Product Images
- Translations
- SEO Metadata
Akeneo PIM
BigCommerce
Salesforce
Distributors
Print / PDF
DAM
⚡ Hot Data Bypass Path
Pricing & Inventory Bypass: D365 F&O → BigCommerce (Direct API) — Real-time pricing and inventory data bypasses Akeneo entirely, flowing directly from ERP to commerce channels. This ensures sub-second price updates and accurate ATP (Available to Promise) without waiting for PIM enrichment cycles.
Data Ownership Model
Clear system-of-record boundaries eliminate sync conflicts and establish governance accountability across the integration.
| Data Domain | System of Record | Sync Direction | Notes |
|---|---|---|---|
| SKU / Item Number | D365 F&O | D365 → Akeneo | Global product identifier, immutable once created |
| Unit of Measure | D365 F&O | D365 → Akeneo | Inventory, purchase, and sales UoM from ERP |
| Pricing / List Price | D365 F&O | D365 → Channels (bypass) | Bypasses Akeneo; flows direct to commerce |
| Inventory / ATP | D365 F&O | D365 → Channels (bypass) | Real-time stock, bypasses Akeneo |
| Cost | D365 F&O | No sync | Internal only, never exposed to PIM or channels |
| Vendor / Supplier | D365 F&O | No sync | Procurement data stays in ERP |
| BOM / Components | D365 F&O | No sync | Bill of materials, ERP internal |
| Marketing Descriptions | Akeneo PIM | Akeneo → Channels | Localizable, channel-scopable rich text |
| Technical Specifications | Akeneo PIM | Akeneo → Channels | Structured attributes per product family |
| Product Images / Assets | Akeneo PIM | Akeneo → Channels | DAM-integrated media management |
| Compliance Documents | Akeneo PIM | Akeneo → Channels | SDS sheets, certifications, regulatory docs |
| Translations | Akeneo PIM | Akeneo → Channels | Multi-locale content management |
| SEO Metadata | Akeneo PIM | Akeneo → Channels | Page titles, meta descriptions, slugs |
| Channel-Specific Content | Akeneo PIM | Akeneo → Channels | Per-channel attribute scoping |
| Product Name | Shared | D365 → Akeneo (seed) → Akeneo enriches | D365 provides initial name; Akeneo may enrich for marketing |
| Category Mapping | Shared | D365 hints → Akeneo assigns | D365 item group informs Akeneo category placement |
| Status / Lifecycle | Shared | Bidirectional awareness | D365 active/inactive; Akeneo enabled/disabled per channel |
Governing Principle: No attribute is authored in both systems. D365 is the source of truth for commercial and transactional data. Akeneo is the source of truth for enrichment, marketing, and channel-specific content. This eliminates sync conflicts and establishes clear governance boundaries across all 70+ operating companies.
Middleware Architecture
Four integration approaches evaluated for the D365-to-Akeneo data bridge, from enterprise-grade Azure services to AI-native MCP orchestration.
Azure Integration Services RECOMMENDED
Logic Apps + Service Bus + Event Grid
Event-driven architecture leveraging Azure's native integration suite. D365 Business Events trigger Logic App workflows that transform and route product data to Akeneo's REST API.
- Azure Service Bus queues for reliable, ordered message delivery with dead-letter handling
- Event Grid for fan-out to multiple consumers (Akeneo, monitoring, analytics)
- Dead Letter Queue captures failed messages for manual review and replay
- Azure Monitor provides end-to-end observability with Application Insights
- Managed connectors for D365 F&O reduce custom code to transformation logic only
This approach is recommended because Heico already operates within the Microsoft ecosystem with D365 F&O across 17+ legal entities. Azure Integration Services provides native connectivity, enterprise-grade reliability, and centralized monitoring.
Custom Node.js Middleware
Express/Fastify on Azure App Service
A custom-built middleware application providing full control over the integration pipeline.
- REST API endpoints receiving D365 webhooks and Business Event payloads
- Custom transformation logic for complex field mappings and data enrichment
- Direct Akeneo REST API calls via the official JavaScript SDK
- Full control over retry policies, circuit breakers, and error handling
- Higher maintenance burden — requires dedicated DevOps and monitoring setup
Best suited as a fallback option if Logic Apps connectors prove insufficient for complex D365 entity relationships or if the team has strong Node.js expertise.
MCP-Based Integration INNOVATIVE
Model Context Protocol Server
An AI-native integration approach using the Model Context Protocol to expose D365 and Akeneo as tool endpoints for AI agent orchestration.
- MCP server exposing D365 and Akeneo as tool endpoints accessible by AI agents
- AI agents orchestrate cross-system workflows (query D365 inventory, enrich in Akeneo)
- Conversational data management — natural language product queries across both systems
- Intelligent routing — AI determines optimal sync paths based on data type and urgency
- Self-healing — AI agents can diagnose and resolve common sync failures
See Section 8 for the full AI and MCP integration architecture.
Akeneo Native Import (Batch)
CSV/XLSX Scheduled Import Jobs
The simplest approach for initial data migration and pilot-phase testing.
- D365 data extracted to CSV via Data Management Framework (DMF) export jobs
- Uploaded to Akeneo via native import profiles with column mapping
- Best for initial data migration and Phase 1 pilot validation
- Auditable — import/export files serve as data snapshots
- No real-time capability — scheduled runs only (hourly, daily)
Recommended for Phase 1 foundation work before investing in real-time infrastructure.
| Criteria | Azure Integration | Custom Node.js | MCP Server | Akeneo Import |
|---|---|---|---|---|
| Real-time capable | ✓ Yes | ✓ Yes | ✓ Yes | ✗ No (batch) |
| Microsoft ecosystem | Native | Partial | Partial | None |
| AI integration | Via Copilot | Manual | Native | None |
| Complexity | Medium | High | Medium | Low |
| Monitoring | Azure Monitor | Custom | Custom + AI | Akeneo logs |
| Best for | Phase 1-2 | Fallback | Phase 3+ | Initial pilot |
D365 F&O Product Data Model
Understanding the D365 product architecture is essential for correct field mapping and multi-company integration design.
Multi-Company Bridge: D365's “Legal Entity” maps directly to Akeneo's “Operating Company” reference entity. When a product is released to a new legal entity in D365, the integration creates or updates the product in Akeneo and assigns the corresponding Operating Company. This is the bridge between ERP's multi-company model and PIM's multi-tenant governance.
Inbound Sync — D365 → Akeneo
Three sync approaches support the phased rollout, from simple batch imports to event-driven real-time updates.
🔌 REST API
Akeneo REST API v7 with OAuth2 authentication. JSON payload transformation from D365 OData entities to Akeneo attribute format.
📄 Batch Import
CSV extraction from D365 Data Management Framework. Uploaded to Akeneo via import profiles. Auditable, simple, and reliable for initial validation.
⚡ Event-Driven
D365 Business Events fire on product create/update. Webhook triggers middleware transformation pipeline. Near-real-time sync to Akeneo API.
Field Mapping Explorer
| D365 Field | D365 Entity | Akeneo Attribute | Type | Transform Notes | Category |
|---|
Delta Sync Strategy: Only changed products are synced. D365 Data Events track modified records. The middleware queries D365's change tracking API (/data/$delta) and pushes only deltas to Akeneo, avoiding full catalog re-import on each cycle. This reduces API calls by 90%+ during steady-state operation.
Outbound Sync — Akeneo → Channels
Enriched product data flows from Akeneo to five distribution channels, each receiving a tailored subset of attributes.
BigCommerce / Commerce.com
Native Akeneo connector. Product listings with structured specifications. AI-driven agentic commerce ready.
Salesforce CRM
Product catalog sync for sales teams. Spec sheets available within opportunity workflows.
Distributor Feeds
EDI, CSV, XML exports tailored to partner data standards and distributor portal requirements.
Print / Spec Sheets
PDF generation for technical data sheets and compliance documentation.
DAM / Marketing
Digital asset distribution for brand portals, campaigns, and web channels.
Channel Attribute Matrix
| Attribute | BigCommerce | Salesforce | Distributor | Print PDF | DAM |
|---|---|---|---|---|---|
| Product Name | ✓ | ✓ | ✓ | ✓ | — |
| Marketing Description | ✓ | ✓ | — | ✓ | — |
| Technical Specs | ✓ | ✓ | ✓ | ✓ | — |
| Product Images | ✓ | — | Thumbnail | ✓ | ✓ Full |
| SEO Title / Meta | ✓ | — | — | — | — |
| Compliance Docs | Link | Link | ✓ | ✓ | ✓ |
| Pricing | Via D365 | Via D365 | Via D365 | — | — |
AI & MCP Integration Layer
The Model Context Protocol enables AI agents to orchestrate cross-system workflows, bridging D365 transactional data with Akeneo enrichment intelligence.
AI Orchestration Pipeline
D365 Data
MCP Server
AI Agent
Akeneo
MCP Tool Catalog
Proposed MCP server tools bridging D365 and Akeneo
D365 MCP Integration: Microsoft's MCP servers for Dynamics 365 expose 650,000+ automatable ERP actions. An AI agent connected to both the D365 MCP server and a custom Akeneo MCP server can orchestrate cross-system workflows: querying inventory in D365, checking enrichment status in Akeneo, and triggering actions in either system — all through a single conversational interface.
AI Enrichment Pipeline
Ingest
D365 pushes new or updated product with technical specifications via Business Events. The middleware captures the payload and queues it for AI processing.
Analyze
AI analyzes product attributes, material data, compliance markers, and existing catalog patterns to understand the product context and identify enrichment opportunities.
Generate
Auto-generate marketing descriptions, SEO metadata, image alt-text, and translation suggestions based on the product's technical data and family patterns.
Validate
Franklin AI checks quality: spell-check, completeness scoring, image quality assessment, and duplicate detection across the catalog before approval.
Publish
Enriched product published to configured channels via Akeneo's export profiles. Channel-specific formatting applied automatically per scope rules.
Franklin AI Capabilities
Automated Categorization
AI suggests product category based on attributes and existing catalog patterns across all product families.
Spell Checking
Multi-language spell checking across all text attributes with context-aware correction suggestions.
Attribute Suggestion
Recommends missing attribute values based on similar products within the same family and category.
Image Quality
Assesses image resolution, background quality, and composition against channel-specific requirements.
Agentic Commerce Vision: Commerce.com's pivot to “agentic commerce” means AI agents will browse, evaluate, and transact on behalf of buyers. Well-structured PIM data — clean attributes, consistent taxonomies, rich descriptions — is the foundation these agents need. The integration architecture proposed here creates the data quality necessary for Heico's products to be discoverable and transactable in an AI-agent-driven commerce future.
Jason Lotzer has deployed a production MCP server on Smithery (94/100 quality score) enabling AI agents to interact with a live commerce platform — demonstrating practical MCP implementation capability.
Security & Governance
Enterprise-grade security controls protecting data integrity across the D365-Akeneo integration boundary.
API Authentication
OAuth 2.0 client credentials flow. Azure AD service principal for D365 access. Akeneo API key with automatic refresh token rotation.
Data Encryption
TLS 1.3 in transit. AES-256 at rest. Azure Key Vault for secret management and certificate rotation.
Audit Logging
Every API call logged with timestamp, service identity, payload hash, and sync result. Retained per compliance policy.
Role-Based API Access
Integration service account with scoped permissions matching Akeneo's user group permission model. Principle of least privilege enforced.
GDPR Compliance
EU subsidiary data handling. Data residency considerations for Wistra (DE), Ancra Systems (NL), TITAN (BR). Right-to-erasure propagation.
Anomaly Detection
Rate limiting on API endpoints. Anomaly detection on sync volumes. Failed auth alerting after 5 consecutive attempts triggers lockout.
Monitoring & Observability
Real-time integration health metrics with automated alerting and escalation paths.
| Metric | Threshold | Alert Level | Action |
|---|---|---|---|
| Sync success rate | <95% | 🔴 Critical | Page on-call, investigate root cause |
| Avg sync latency | >10s | 🟠 Warning | Check middleware throughput |
| Dead letter queue | >100 messages | 🔴 Critical | Manual review required |
| Data completeness | <80% per family | 🟠 Warning | Enrichment queue review |
| API auth failures | >5 in 1 hour | 🔴 Critical | Security review, credential rotation |
| Channel export errors | >2% | 🟠 Warning | Connector health check |
Axiamatic Integration: Integration health metrics feed directly into Axiamatic's AI-powered program governance dashboard, providing CIO Thomas Gerdes and the IT leadership team with real-time risk visibility and delivery predictability — consistent with how all 20+ active technology programs at Heico are managed.
Implementation Roadmap
Phased rollout building from validated batch import through real-time integration to AI-native enrichment and agentic commerce readiness.
Batch Sync & Pilot Validation
Establish core D365 → Akeneo sync using batch CSV import. Pilot with ASG Cargo Solutions (Kinedyne, Ancra) and Electronic Passives (Ohmite, Arcol). Validate field mappings, data quality thresholds, and Operating Company governance model.
Event-Driven Sync & Channel Activation
Transition from batch to near-real-time event-driven sync. Deploy Azure Integration Services middleware. Activate outbound channel connectors for BigCommerce and Salesforce CRM. Establish monitoring dashboard and alerting.
AI-Native Enrichment & Cross-System Intelligence
Deploy MCP server bridging D365 and Akeneo. Activate AI-powered enrichment pipeline for automated description generation, category suggestion, and data quality validation. Enable Franklin AI for native Akeneo intelligence.
Agentic Commerce & Acquisition Playbook
Enable AI agent product discovery for agentic commerce. Build repeatable acquisition onboarding playbook — each new Heico acquisition gets PIM-ready within weeks, not months. Expand AI enrichment to multi-language generation and advanced analytics.