Your ERP is only as useful as the structure it sits on, and that structure was designed for the business you were, not the one you became.
A company invests heavily in Oracle Fusion. The implementation takes about a year and goes live on the enterprise structure agreed in the first workshops. Then the business grows. A few acquisitions follow, and each new entity is fitted into the existing structure because that is the quickest way to get it to be transacted.
Five years later, the CEO asks: “What is the profitability of each of our lines of business?” The data is all in the ERP. But the lines of business are buried so deep in a structure designed years ago that no standard report can pull them apart. So, the answer becomes: “Give us a couple of weeks.”
When the structure does not match how management wants to see the business, the ERP quietly taxes your people. Managers export data and rebuild the view they need in spreadsheets, every month, quarter, and year. The pain spreads further:
All this despite an ERP investment worth millions. It is why the enterprise structure deserves real thought at the start, and why fixing it later is worth the effort.
No. If the requirements are clear and pain points are identified, enterprise restructuring typically takes two to three months, not another 9-10 months implementation cycle. Two things make that possible:
We delivered quite a few such restructurings on a live Oracle Fusion environment in about two to three months, and leadership now gets line-of-business profitability straight from the system.
Your business team already knows the pain points: the report that takes three days, the allocation nobody trusts, the transfer whose margin nobody can see. Trace each one back to the level of the structure that causes it: Enterprise, Ledger, Legal Entity, Business Unit, Inventory Organization or Cost Organization Structure.
This matters because the deeper the level you change, the more objects you touch. A tweak at inventory organization level is contained. A change at ledger or legal entity level ripples through setups, balances, subledger accounting, approvals, integrations and reports. Knowing the level up front is what keeps the effort at two to three months.
Keep it a structure change
One principle protects this type of niche project: this is an enterprise structure change, not a business process change. Holding this line keeps training light, resistance low and scope under control.
Then ask the basic questions honestly:
There is no universal right answer. The right answer is the one that lets leadership see profitability the way they think about the business, while still meeting statutory needs. And hold one principle: this is an enterprise structure change, not a business process change, which keeps training light and resistance low.
If goods move between lines of business at a profit, consider enabling Financial Orchestration. It tracks profitability across intercompany and internal transfers, so each line of business shows what it really earns instead of leaving that margin hidden inside the group. A restructuring is also a good moment to revisit lot uniqueness. If it was never defined during the initial configuration, define it now, so lot-controlled items stay traceable and consistent across the new structure.
The data already lives in the ERP, so extraction is the easy part. At Orbrick, we use tailored reports that extract the data directly in the Oracle uploader format. The business can then work on it in that same format, and the SI can load it with minimal rework. The real work is converting the data to fit the new structure, whether the structure is being expanded (one entity becomes several) or clubbed (several become one). Work in three layers: configuration (approvals, Subledger Accounting and key module setups), then master data, then transactional data.
Start with a clean cut. Close open transactions, bring clearing accounts to zero, and reconcile balances such as cost of goods sold and accruals against the pending transactions behind them. Balances nobody can explain in the old structure get harder to explain in the new one.
Choose tools by volume: FBDI for large master and transactional data, Functional Setup Manager for configuration, ADFDI and Macros for smaller loads, and the Oracle VBS extension in Excel for objects that have no FBDI or FSM template but do have an API.
Then rehearse. Run at least two full cycles: a process playback to expose gaps in templates, conversion rules and configuration, and a production run timed step by step, so the cutover plan rests on measured durations. Reconcile every cycle two ways: technically (records loaded versus records signed off) and by business (subledger totals tie back to control accounts). Data quality is the business’s responsibility, and for UAT, migrate 100% of the data, because samples hide the problems that only appear at real volume.
Replace old-structure conditions in every integration mapping. For reports, create new versions restricted to the new structure rather than editing in place. Performance stays healthy, old and new data never appears side by side, and the old reports remain available to the few people who need to reconcile.
Start with communication, not test scripts. Hearing “enterprise restructuring”, many users assume downsizing. Say plainly: this restructures how the system is organized, not the organization. Roles and day-to-day work stay the same. Then have every user run their routine transactions and the reports they depend on, so the structure is proven on real work before it reaches production.
Old and new structures coexist for a while. Everyday users with access to both may post to the old one by habit and create rework, while finance leaders and a few key users need both for the first months to reconcile. Decide access role by role: most users get the new structure only, and a small named group gets both, with clear rules on who may transact and who may only view.
Cutover needs a tightly governed plan. After the cutover date there is a blackout period when the business holds off creating transactions, so users must know how to record them meanwhile and how they will be loaded after go-live. Verify and reconcile every load, and build the plan on timings measured in rehearsal, not on hope.
Reconcile again against the migration strategy, clear the blackout backlog module by module, evaluate processes and data continuously, and monitor the impact: time spent on reporting, speed of the close, and visibility of profitability by line of business.
An ERP does not stay right just because it went live right. Businesses acquire, reorganize and change how they measure themselves, and the structure must keep pace, or your teams will make up the difference by hand.
A well-restructured ERP gives you more than one better report. It gives you a structure that can welcome the next acquisition, the next line of business and the next tough question from the boardroom, without another round of manual workarounds. The best time to get the structure right was before go-live. The next best time is now.
Shankar is an SCM consultant who is expert in distribution, this person combines a passion for anime and fluency in Japanese with a love for fitness. His expertise lies in optimizing supply chains while staying active and engaged with Japanese culture.
Actionable insights on ERP, AI and business solutions, straight to your inbox each month.
Subscribe for InsightsPick Orbrick as a preferred source and Google is more likely to show you our insights in Top Stories and AI Mode.
January 16, 2024
January 16, 2024
Be part of our growing community. Subscribe to our monthly newsletter and get actionable insights on ERP, AI, business solutions to optimize your ongoing operations
Subscribe for InsightsBegin your Business Value Maximization journey with us. Schedule a complimentary consultation today to understand how we make it a smooth ride for you.
Contact Us