What is Planning?

All organizations perform activities across functional domains and industry verticals. Teams set targets for subsequent years based on historical data, current performance, market trajectory, and a variety of other factors. Budget, forecast, and stretch values are calculated and published. Following that, the budget and forecast are compared to actuals on a monthly basis, and performance is tracked. If necessary, values are re-adjusted once approvals are received.

SAP Analytics Cloud – Planning

SAC Planning models facilitate generating collaborative Budgets, Forecasts and integrate with realized actuals from enterprise systems on a near real-time basis. Planning and forecasting accuracy can be improved with advanced features such as Variance management, Value Driver Tree, Predictive Modeling, and integration with R.

Data Manipulation

Data can be loaded only directly into a Dimension, whereas data can be loaded into a Model or copied from another. Note: In this article, a quick snapshot of data load behavior is provided. This article focuses on planning models predominantly, but analytical models also have overlapping concepts. A copy action facilitates moving data from one model to another and there are two types. Here, the Version is the mandatory dimension that needs to be updated at all times
  • Copy Step: This step copies data within the model. This is always between two versions due to version dimensions. Using copy rules, data can be filtered, aggregated, and cross-mapped. The hierarchy will be applied to leaf members automatically.
  • Cross-Model Copy Step: Allows users to transfer data between different models. Cardinality plays a key role and conformed public dimensions are automatically matched. Data can be cross-mapped and filtered. Aggregation is controlled by the definition maintained in measures within the model and is not available as a separate option.
  • Append vs Overwrite: Append implies that the measure is aggregated along existing values i.e., cumulative in nature. Overwrite replaces the old value with the new value. This setting is only for matching cardinality records and any new data is automatically inserted i.e., appended.
  • Points to be aware of
    • Runtime formula cannot be used in copy actions
    • Some calculated measures can be copied from one model to another and needs to be checked for performance impact
    • Cross-dimensional mapping is possible .e.g. Sending Cost Center to Receiving Cost center
    • Data is unbooked by default
    • Target version is a mandatory parameter and choosing source and target as the same version will cause data inconsistency
    • Aggregation feature works similarly to GROUP BY/HAVING clause in SQL
No Type Use Case Observation
1 Copy Simple copy within the model. Model with 1 Local and 2 Public dimensions with 2 Measures. Default mapping generated is applied as-is. Data is summarized based on member values at the lowest granularity for all dimensions under aggregation. Acts as an activity to copy the dataset to a different version without any changes
2 Copy Simple copy within the model and aggregation along a dimension. Model with 1 Local and 2 Public dimensions with 2 Measures. Copy data within the model with aggregation for 1 public dimension member. All dimensions except the aggregate dimension have the granularity maintained intact. Measures aggregated to the specified dimensional value in the target version.
3 Copy Simple copy within the model and aggregation along multiple dimensions. Model with multiple dimensions and  Measures. Copy data within the model with aggregation for 2 or more dimensions Adding additional dimensions for aggregation results in result rows being summarized according to the members selected for ‘ALL’ dimensions and the granularity is adjusted according to cardinality in the target dimension
3 Copy Simple copy within the model with aggregation along the hierarchy Standard model with standard dimension, measures, and one public dimension with a hierarchy e.g. COUNTRY and STATE with the state being leaf member nodes. Hierarchy is ignored and only leaf members can be aggregated. Cardinality and target version constraints apply as-is.
4 Cross Copy Date/Time dimension behavior. Copy data from Model A to Model B with all dimensions being the same except Model A has Date granularity and Model B has year granularity. Data is auto-aggregated based on target granularity. Note: Custom mapping can be generated based on source hierarchy e.g. FY-FQ-FP can be used to map against CY-CQ-CM
5 Cross Copy Mismatched cardinality from source to target models. Copy data from Model A which has 3 public dimensions to Model B that 4 public dimensions The default value must be assigned to the target missing dimension. Data automatically aggregates to the default value. Hierarchy cannot be used due to the complexity of disaggregation and only leaf members can be selected.
6 Cross Copy Mismatched cardinality from source to target models. Copy data from Model A that has 4 public dimensions to Model B that 3 public dimensions Dimension/ entity that is not part of the target is ignored during the copy. Measures are automatically aggregated to target cardinality.

Authors

  • Vignesh M
  • Karpagam K.