Allocation Processes, Steps, and Rules

For each planning-enabled model, you can create multiple allocation processes in the Allocation window, and define each process by adding allocation steps and rules.

Allocation processes can consist of a single step, or a sequence of steps. You execute the process in a table to perform its allocation steps on the data for that table.

An allocation step as a whole determines how values from one source dimension is distributed among members of target dimensions. (It's possible to have multiple target dimensions.) You can also run allocation steps in data actions. See Data Actions for details.

The allocations for a step are governed by allocation rules. An allocation rule specifies that values in source dimension members are to be allocated to target dimension members using a driver. A driver is a variable, such as the level of activity or volume, that causally affects costs over a given time span. That is, there is a cause-and-effect relationship between a change in the level of activity or volume and a change in the level of total costs. For example, changes in the driver headcount affect salaries, or changes in square meters of floorspace affect rent. You can also choose the Direct Assignment option, which assigns all of the source member's value to a single target member.

Working with allocation steps

You can create an allocation step as part of a process, or create the step on its own and add it to a process or data action later.

To create an allocation step, you must specify a source dimension, which holds the values that you want to distribute. Then you specify a target dimension to receive the distributed values. You can also specify an overwrite dimension, which will allow you to create overrides for allocation rules. An override is a more specific allocation rule that takes preference over a more general one. For example, you could create an allocation rule that the travel costs for all cost centers is allocated to all products using the driver Revenue, except for the exception (the override) that for the United States cost center, the travel costs are directly assigned to the single product, Laptop.

Note

Allocation rules are executed in the order of the more specific first, not in the order in which you create them.

It is possible to use the same dimension as the source and target for an allocation step. In this case, values will be reallocated from a source member to one or more members of the same dimension hierarchy. For example, you can use reallocations to take costs that have already been assigned to a software support department, and reassign them to all of the departments in your organization based on the number of support hours that they used. This is known as the reciprocal method of cost allocation. Since some of the original value may be allocated back to the source member, you can choose to perform multiple repetitions of the step to further reduce this amount.

You can also specify reference dimensions for an allocation step. Reference dimensions let you use more fine-grained weights for the allocation because only weights matching a record to be allocated in all reference dimensions will be used for its allocation. Allocation steps automatically use the Date dimension as a reference dimension. This breaks the values down by time. For example, assume all Salaries are allocated to all Cost Centers based on Headcount, and the Date dimension uses month granularity. This means the Salaries costs will be broken down to the month level based on the Headcount specified per month, and then assigned to the Cost Centers. You can add more reference dimensions if you need to break the values down further.

Note

Using more than one reference dimension can slow down your allocation process. For better performance, try filtering your reference dimensions when you run the allocation.

You may also want to exclude some accounts from the allocation step. For example, you want to allocate values for all financial accounts, but leave driver accounts unchanged. In this case, you can use the Filter by Accounts feature to select only the financial accounts.

Finally, you may need to use one set of accounts to calculate the values to allocate, and then book those values to a different account. To do this, you can select a Booking Account to receive the values. All changes in value for the allocation step, both for source members and target members, are carried out on the booking account. You can use the Keep Source setting if you don’t want to change the value of the booking account's source members, and the Overwrite Target setting if you want to overwrite the values of the target cells.

Adding rules to a step

Once you have created an allocation step, you can begin to add allocation rules to the new step. To create an allocation rule, you select a member of the source dimension, choose a driver (or direct assignment), and choose a member or members of the target dimension. If you have added an overwrite dimension, you can add an override.

If you need to pick driver values from a specific version, you can select a calculated account that uses a LOOKUP function to refer to that version. For more information, see Lookup and Reference Functions.

You can also create multiple rules by copying and pasting this data directly into a table of rules for the step.

Working with processes

As you create steps, you can visualize the entire allocation process in the Steps Overview pane, which can be displayed or hidden by selecting (Details). To rearrange the step order, select and drag steps on the tab bar of the allocation process page. You can also load existing steps based on the same model into your process.

Creating a sequence of multiple steps allows you to perform complex allocations, for example, allocating costs from accounts to different projects and products, and then further allocating costs from those dimensions to different geographical responsibility centers.

Once the allocation process is complete and saved, you can open a table to execute the process.