3. Business Drivers

​

This manufacturing company had multiple compelling drivers:

​

1. Legacy Technical Debt & High Maintenance Costs: Up to 70% of the IT budget was tied to legacy application maintenance—including mainframes and outdated ERP modules. These systems used bespoke code, obsolete databases, and hardware near end-of-life.

2. Operational Inefficiency & Duplication: Engineering teams used older BIM software on local servers; CAD files were stored on regional file shares with versioning issues. Lack of centralized document control slowed tendering and fabrication prep across plants.

3. Scalability Limits: The manufacturing company’s international expansion strained capacity—time-consuming system provisioning and ad hoc departmental servers limited new plant ramp-ups.

4. Fragmented Collaboration: Email and file-sharing tools varied by region; engineers, supply chain, and finance teams lacked synchronized channels, incurring daily overhead.

5. Competitive Pressure & Innovation Needs: With the rise of digital construction, generative AI for weld inspection, and advanced data analytics.

​

The shift to cloud offered an opportunity to reduce costs, scale operations rapidly, consolidate collaboration, and introduce advanced data analytics and automation.

​

4. Current Legacy Landscape

​

Before modernization, the manufacturing company’s IT landscape included:

​

• On‑prem ERP based on decade‑old proprietary modules, patched over time.

• Regional CAD/BIM file servers, often using unsupported OS, with no unified metadata taxonomy.

• Email domains, conferencing, and collaboration tools varied by region; outdated antivirus software and no centralized user directory.

• Isolated databases for scheduling, quality control, and sub‑contract management, with limited integration and high latency.

• No cloud-native data analytics; most BI dashboards are refreshed offline with manual extract scripts.

 

Many systems had no vendor support and relied on internal “institutional knowledge.” Finely-tuned yet fragile workflows meant any disruption risked delays, but nonetheless, the systems were brittle and posed a growing impediment to expansion and innovation.

​

5. Pilot Phase & Cloud Platform Selection

​

The EA team oversaw a two-month pilot deployment centered on Google Workspace. Approximately 250 employees across finance, engineering, and admin were trained on Gmail, Drive, Docs/Sheets/Slides, and Meet features, along with beta access to AI-enhanced tools.

​

This was followed by architecture design sessions addressing:

​

• BIM file archiving and collaboration are scalable via Google Drive + Vault.

• Retiring legacy SharePoint and regional file servers.

• Cloud data lake on Azure/AWS for ERP, sensor data, and reporting pipelines.

• Identity and access revised to Google Directory and Azure AD sync.

​

These pilots surfaced real-world lessons: need for metadata cleanup, offline access planning, and network capacity upgrades before global roll-out.

​

6. EA Workshop Feedback

​

After a comprehensive 3-day workshop, the EAs shared candid feedback:

​

“Thank you for sharing knowledge. I learned new concepts that we will certainly explore in the short term.”

“Interesting training that will help me focus on the important things to deliver value in my work as an enterprise architect.”

​

Their responses reflect both appreciation and readiness to apply new techniques such as capability‑based and value stream mapping, and the integration of enterprise architecture with an agile practice.

​

7. Outcomes & Business Value

​

Early quantitative and qualitative outcomes:

​

• Cost reduction: Designated legacy maintenance spending dropped by 20% within the first year post‑pilot. Infrastructure refresh and license consolidation saved $350K in Year 1.

• Operational efficiency: Employee productivity via collaborative documents increased 20%, and synchronized BIM workflows reduced model errors by almost 15%.

• Scalability & resilience: A new plant was provisioned in 6 weeks instead of 4 months. Onboarding new users was streamlined via cloud directory sync.

• Security & governance: Google Workspace’s built‑in DLP, admin insights, and data retention policies provided immediate improvement in compliance readiness.

• Innovation capacity: The enterprise data lake enabled the first use cases for generative AI, like weld inspection and supplier lead‑time forecasting.

​

8. Risks, Challenges & Mitigations

​

Key challenges and how the EA team managed them:

​

1. Data Quality and Mapping: Legacy data required cleansing and field remapping. EA set up a small ETL squad to apply data validation rules before migration.

2. Integration Complexity: ERP–BIM–schedule interdependencies posed timing risks. The team used micro‑batch processing and API wrappers for phased cut‑over.

3. User Adoption Resistance: Longtime power users of old systems hesitated. The EA included “Google Guides” in each division to support peers and reinforce day‑to‑day validation.

4. Vendor Lock‑In and Cost Control: EA developed a cloud cost model and implemented FinOps discipline, refactoring services to avoid vendor-specific proprietary functions.

​

9. Lessons Learned

​

• Do not rush “lift‑and‑shift”: the team found that rearchitecting core apps (for example, ERP module) upfront yielded better ROI than deferring refactoring. This aligns with modern EA best practices.

• Prioritize enterprise data governance: The data architecture decisions during the pilot laid the foundation for later AI and advanced analytics.

• Balance process with flexibility: A lightweight EA governance board minimized red tape while ensuring cross‑module consistency.

• Invest in people early: Training early adopters built internal champions, lessening resistance through peer-to-peer support.

​

10. Conclusion

​

This manufacturing company’s cloud migration and legacy modernization use case exemplifies a pragmatic, business-driven EA transformation. The five-person EA team employed workshop‑based design, pilot deployments, and structured cost modeling to drive measurable business value. The candid feedback from the EAs, an eagerness to learn, and pragmatism in execution were central to success. No sugar-coating: migrating legacy systems is complex work. But with disciplined architecture governance, phased delivery, and a solid data strategy, this manufacturing company built a resilient, scalable foundation for its digital future.