Hello, I'm working on a test to get T&E data that is pulling from an input module into the "Collector" module(using a line item subset) to then map to the actual accounts in a P&L. The collector module has two line items Data (which is Collect) and mapping. I formatted the mapping item as the "Accounts" list, but I wasnt sure if this was the correct way of doing it. I know there are a number of workaround ways like using boolean checks and formulas through more modules but the goal here is to learn Anaplan best practice and eliminate any unnecsary calcs so any feedback would be appreciated. I included a screenshot of the collector module.
I've often used similar solution and have found it very useful. I usually use separate modules for mappings. Let's call mapping module "Data to Account Mapping". It would contain only one dimension (which is the line item subset that is also in the data collector module) and one line item "Account" that would be list-formatted with your account list. Time and Version would not be included here.
Then you could use this kind of formula in your P&L:
Data Collector Module.Data[SUM: Data to Account Mapping.Account]
Another option here would be to put the mapping line item directly in the input module, and then you don't have to worry about the line item subset. The way that this would work is to have your input line item as well as a reference line item, in the same module, formatted to the accounts list and referencing, by formula, the member of your accounts list that is the target to include in your financial statement. Then you would be able to consolidate your individually planned accounts into a P&L module utilizing the formula not unlike the one previously commented, T&E Module.Input Line Item[SUM:Mapping Line Item]. This solution is, from my experience, best practice because it minimizes model size and streamlines the aggregation (you don't have to have a separate "staging" module containing your line item subset), provides ease of use for the model builder, since the reference is with the input, and you don't add the complexity of having to create many line item subsets (or manage the subset to contain all of the different module's line items) to do the mapping.
I agree that Ethan's suggestion is another option to consider. However, does it really minimize the model size? Doesn't having separate line items for account in the input module mean that you will double the number of (leaf level) cells for those line items that you add account information to? If you build a separate "collector module" with the same dimensions, you will also double the number of leaf level cells (assuming that aggregation is set to none). If you need to collect the data with less dimensions or at a higher level than input, using collector module will be smaller in size.
If you need to optimize the number of cells, you'll probably create an assumptions module with account formatted line item and point to that in the formula: Input module.Source line item A[SUM: Assumptions module.Account formatted line item A] + Input module.Source line item B[SUM: Assumptions module.Account formatted line item B]... This will not duplicate the number of cells and you will get the figures to accounts. What I dislike in this approach is that formulas will quickly become super long and difficult to interpret/maintain. You need to sum all the line items in the formula (which sometimes might be a very large number of line items). That's why I have liked collector modules where you only need to include new line items to the line item subset and add an account mapping in the mapping module. No need to touch any formulas.