In Oracle EBS (E-Business Suite), divestitures or carve-outs can be performed using three main approaches: Top-Down Divestiture, Divesture by Balancing Segment Value (BSV), and Bottom-Up Divestiture. The key difference lies in the complexity of separating data, which directly impacts the time and cost of the process.  The purpose of a divestiture is to separate part of a company so that it can be sold.  Since the carved-out piece of the business may be sold to a competitor, an important component of this separation is identifying the data that the separated business needs to run on its own apart from the parent company, while keeping the data of the parent company confidential.  For purposes of this paper, the separated environment is called the child instance.

Software works by a clone and purge approach, identifying the basic carve-out criteria and then tracing the data throughout the environment to find all related data.  The parent instance is cloned, then data that is not needed for the child instance is purged from the instance.  Data that is co-mingled or split between the parent and child adds to the complexity of the divestiture and requires business decisions to determine what data is retained and how the related transactions are processed to maintain the integrity of the database.

Top-Down Divestiture

This approach is simpler and more efficient because it leverages the natural, hierarchical structure of Oracle EBS.

• Data Hierarchy: Oracle EBS organizes data logically in a hierarchy, starting with the Ledger and its associated Operating Units. These are “natural separators” of data, as user access and security are already defined at these levels. This hierarchy provides a Top-Down delineation of data, transactions, workflows, metadata, etc. which can used by software to more easily perform the data separations and purges needed to create the desired child instance.

Divesture by Balancing Segment Value (BSV) or Legal Entity

• Within the data hierarchy, the Legal Entity or BSV is another financial reporting designator that helps balance financial entries (ensuring each financial entry has equal debit and credits). While entries are coded with a BSV to allow reporting aggregation, the data isn’t physically separated and the complexity stems from the fact that while transactions are tagged with a BSV, it is instead commingled within thousands of interdependent tables.  The BSV can also be used as a starting point to identify and separate the necessary data.
• While the BSV is not the highest level of the data hierarchy, it is a defined and integral part of the financial and legal structure. It serves as a “natural separator” for legal and financial reporting, allowing for complete trial balance generation. This means that even with a BSV-based divestiture, the system must perform complex data filtering, identify dependencies, and resolve all financial balances.

Bottom-Up Divestiture

This approach is more complex and time-consuming because it attempts to separate data based on criteria that are not part of the established Oracle EBS hierarchy.

• Criteria: Bottom-up divestitures use lower-level criteria for data separation, such as a product line or a value within a non-balancing segment of the chart of accounts.
• Process: Unlike the top-down approach, these criteria have no direct relationship to the top of the EBS hierarchy or BSVs. As a result, transactions must be identified and mapped out individually across various EBS modules. This “detection process” is highly complex and results in many commingled transactions that require resolution.
• Complexity: The software needs thousands of additional rules about relationships to separate this commingled data. This increases the risk of “broken dependencies,” that require business decisions about what data to retain and purge for the associated, dependent data. The process also requires recalculating balances after the data is split.

Complexity, Cost, and Conclusion:

Because of the thousands of rules involved in identifying and validating data for a carve-out (even a top-down approach), divestures must be done with software that can manage relationships and joins among potentially thousands of tables among different types of data and still maintain data integrity and a fully operational environment for the child company.  Top-Down divestitures generally take less time and are less expensive than Bottom-up divestitures.  Depending on the environment, the modules installed, the number of comingled transactions, and the level of carve-out criteria, a Bottom-Up carve-out requires a multifaceted approach of discovering and mapping transactions, which can make it take up to twice as long and cost more. Divestitures by BSV or Legal Entity generally fall somewhere in the middle. In conclusion, when determining how to approach your divestiture, it is critical to understand the carve out criteria, and make an informed decision based on how your data is organized, taking into consideration the data complexity, timeline, and budget for your project.