Oracle Cloud ERP 22D introduces the new Cross-Validation Combination Sets feature. Combination Sets work with Related Value Sets and Cross Validation Rules to ensure that account combinations comply with your accounting policies. In this post, we contrast the three features and offer some guidance for their combined use.

The Chart of Accounts, Segment Values and Account Combinations

The Cloud ERP chart of accounts (COA) is a foundational enterprise structure. See blog post Oracle Cloud ERP Chart of Accounts Design Considerations for key considerations and best practices.

We recommend that you define and manage COA segments, values and hierarchies in Enterprise Data Management (EDMCS). When you create a new segment value in EDMCS, it is automatically propagated to Cloud ERP. For more details, see EDMCS’s Integrating with Oracle Financials Cloud General Ledger.

This blog focuses on how to control the creation of new combinations of segment values (aka account combinations). For example, how to ensure that no expenses are attributed to a legal entity which never participates in a cost center’s activity.

A key design decision is whether to enable dynamic insertion in Cloud ERP.

If you disable dynamic insertion, an external process will manage the process of account combination validation. When it approves a new account combination, use the Import Account Combinations FBDI to import it into Cloud ERP.

However, by far the majority of customers enable dynamic insertion. They control which combinations of segments values are valid using ERP features: Related Value Sets, Cross Validation Rules and Combination Sets. Upstream external applications that generate new account combinations can validate them using the Account Combination Validation web service.

Commonalities Between Account Combination Validation Features

Combination Sets, Related Value Sets, and Cross Validation Rules validate new account combinations in Cloud ERP.  Account combinations created before a validation rule has been activated are ignored. If you deploy a rule that invalidates an existing account combination, you must actively disable the combination to prevent its future use. Account combinations can be disabled:

All three validation rules enforce validation irrespective of whether the UI or automated processing creates the account combination.

Combination Sets, Related Value Sets and Cross Validation Rules validate account combinations irrespective of which user creates or views them. To address requirements to secure account combinations depending on a user’s role, use Segment Value Security or Data Access Sets.

Related Value Sets

Related Value Sets define a simple relationship between two segments of the accounting flexfield. For each segment value in the driving segment, you define valid dependent segment values. Related Value Sets support:

  • segments assigned independent value sets.
  • dependent values assigned to multiple driving segment values.
  • child (leaf node) segment values only.
  • a direct comparison between segment values. No ranges, wildcards, or tree operators are supported.

Example of Related Value Sets

When a user enters an account combination through the UI, Related Value Sets displays only valid dependent segment values. The driver segment must therefore precede the dependent segment in your chart of accounts definition. Once two value sets are related, you must map every value in the driver segment to one or more values in the dependent segment.

You can chain Related Value Sets together. For example, the Cost Center value set could depend on the Company value set; and the Location value set depend on the Cost Center value set.

You create Related Value Sets through the UI or a spreadsheet. The spreadsheet upload supports incremental loads only. FBDI is not available for Related Value Sets.

Use Related Value Sets:

  • when a simple listing of valid pairs of values characterizes the relationship between the two segments.
  • when there is a requirement to display only valid dependent segment values in the UI.

Avoid Related Value Sets:

  • when the total number of valid pairs of values runs into many thousands.
  • when the relationship between the driver and dependent segment values applies only to a small subset of driver segment values.
  • when you define and maintain your account combination validation rules in an external master data management (MDM) application or Enterprise Data Management (EDMCS).

Cross Validation Rules

Cross Validation Rules define a complex relationship between segment values in two or more segments. Each rule consists of a condition and validation filter. The condition filter identifies which combinations of segment values are subject to the rule. The validation filter defines which dependent segment values are valid to combine with them.

Manage Cross Validations Rules page

Cross Validation Rules offer considerable flexibility for both filters. The ability to use account hierarchies (trees), for example, simplifies rules and minimizes maintenance since many changes to a hierarchy will not impact Cross Validation Rules that use it.

You create Cross Validation Rules through the UI or FBDI which supports replace mode. FBDI offers spreadsheet entry or automated integration orchestrated by a web service.

Use Cross Validation Rules:

  • when the relationship between segment values applies to only a subset of the condition’s values. For example, for a few Companies, there are a limited number of valid Cost Centers. For the majority of Companies, all Cost Centers are valid.
  • when a pre-existing segment value hierarchy reflects the relationship between segment values. For example, one or more parents in the Location hierarchy maps to a Company segment value. Note that creating a hierarchy solely for the purpose of validating account combinations may be counterproductive.
  • when a range reflects the relationship between segment values. For example, Location segment values between 1000 and 2000 map to a Cost Center segment value.

Avoid Cross validation rules:

  • when the relationship between two segments requires thousands of rules to define. Creating a new combination evaluates each Cross Validation Rule. A huge number substantially increases the time it takes to create new account combinations.
  • when you wish to maintain your account combination validation rules in EDMCS or other external MDM that does not support complex validation rules.

Cross-Validation Combinations

Cross-Validation Combination Sets define up to five validation segments whose segment value combinations are allowed or disallowed. The validation segment order should mirror the chart of accounts definition.

For each allow Combination Set, you must define all valid (partial) combinations of segment values. Other combinations are considered invalid. Conversely, all undefined combinations for a disallow Combination Set are considered valid. For example, a allow Combination Set of Company and Cost Center segment values, implicitly invalidates all other combinations of the two segments’ values.

You create Combination Sets through the UI.

Manage Cross-Validation Combination Sets page

You maintain combinations through FBDI which supports replace mode. FBDI offers spreadsheet entry and automated integration orchestrated by a web service.

Cross-Validation Combinations FBDI

Use Combination Sets:

  • when account combination business rules don’t follow discernible patterns. There are just lists of valid or invalid combinations.
  • when you define and maintain your account combination validation rules in an MDM which supports list-based business rules. EDMCS supports Combination Sets through its custom model feature.
  • when the business rule means that only a few (partial) combinations are valid. A limited number of disallow Combination Set rows will ease maintenance and improve performance.

Avoid Combination Sets:

  • when a pre-existing segment value hierarchy reflects the relationship between segment values.
  • when a range reflects the relationship between segment values.
  • when the business rule depends on more than five segments. Our analysis shows that this rarely occurs.

In all three cases, cross validation rules are more appropriate.

Working Together

Related Value Sets, Cross Validation Rule and Combination Sets work together to validate new account combinations. Here are a few details about how they interact.

Invalid account combinations are always prevented. If a Combination Set conflicts with another Combination Set, Cross Validation Rule, or Related Value Set then disallow rules will always be respected. For example, if a disallow Combination Set specifies a partial combination that a Cross Validation Rule allows, the new account combination will fail validation.

The order in which Cloud ERP evaluates validation rules determines which message displays in the UI and log files:

  • In the UI: Related Value Sets, Combination Sets and then Cross Validation Rules.
  • In automated processing: Combination Sets, Related Value Sets and then Cross Validation Rules.

22D also includes an account combination Validate feature. You can test how your Combination Sets, Related Value Sets and Cross Validation Rules work together from both the Combination Sets and Cross Validation Rules pages. The Validate feature does not save tested account combinations in the database.

We encourage you to implement EDMCS to define and maintain accounting flexfield value sets. For more information about EDMCS and its integration with Cloud ERP, see Integrating Oracle ERP Cloud Oracle General Ledger Applications.

When There is a Choice About Which One to Use

Consider the following factors.

EDMCS integration

If you plan to use EDMCS to define and manage your accounting flexfield validation rules, use Combination Sets.

Ease of Maintenance

Clearly fewer rules reduce the time and effort required to maintain them. Note that defining a hierarchy solely to reduce the number of cross validation rules may be counterproductive. The maintenance effort will simply shift from validation rules to tree maintenance.

Define rules that are stable over time. For example, Cross Validation Rules based on account hierarchies reduce maintenance if the parent is rarely impacted by hierarchy reorganizations.

Performance

Fewer rules improves performance. For example, a single Cross Validation Rule validation filter that specifies a range of segment values will usually be more efficient than many Related Value Sets or Combination Sets.

Avoid overlapping ‘allow’ rules since they are always checked. For example, if a Related Value Set and allow Combination Set are defined for the same segments and values, they will both be checked. The same applies to allow Combination Sets and widely applicable Cross Validation Rule condition filters. On the other hand, processing stops as soon as an invalid account combination has been encountered. For example, if a new account combination fails validation due to a Combination Set, Cross Validation Rules will not be evaluated.

Generally speaking, Combination Sets will perform better than Cross Validation Rules that have a similar number of condition and validation filters.

As always, we recommend you thoroughly test your accounting flexfield validation rules on an appropriately sized environment with realistic data volumes.

Conclusion

Combination Sets add a powerful new method to validate your new account combinations. In many cases, they will simplify rule maintenance and improve performance. However, there is a still a place for Related Value Sets and Cross Validation Rules depending on your business requirements.