I have a hierarchical list of Customers where the bottom two levels are Customer and Branch. I've got a separate hierarchy for products which as its bottom level has SKU. Due to space considerations our Anaplan consultant created a level below Branch for SKU, where items are only created where there is data. We've got actions that import transactions into a LOAD list, then a DATA module in terms of that LOAD list with a view that facilitates creating Branch->SKU list items when they exist in the data.
We now have the requirement that some forecast data is loaded in at the Customer and SKU level then broken down into Branches (based on actual data which is by Branch/SKU). Due to size issues I'm planning on creating a level below Customer which has the SKU, so we'd end up with; "P6 SKU", "C8 CustId", "C9 Branch", "C10 SKU" by Branch and now "C9 SKU by CustId". I can create the C9 SKU by CustId items using the same method as C10, however I'm trying get my head around how to handle breaking these figures down by branch, when the C10 items relating to each branch/SKU may not have been created for a given Customer/SKU.
The step I think I'm missing would be a module that enables creation of the missing Branch/SKU C10 list items, however I'm not sure how to construct that. I'm new to Anaplan, and have only seen List item creation from a module that uses a LOAD list.
Am I thinking about this in the right way? How can I get the additional C10 list items created so I can end up with a module that can do the spreading of Customer/SKU to Branch/SKU?
My understanding is that the Create action is what I would use if I want users to be able to create list items one at a time, however in my use case the forecast comes from an external import, hence I can't expect users to manually add every combination of Branch and SKU applicable (based on the imported file's CustId/SKU combinations together with the historic actuals Branch/SKU split). Is there some way of automating their creation using Actions? Or is there some other method I need? The main case to handle would be where historic data doesn't contain any Branch/SKU list item because we've never sold the SKU to any customer's branch, but according to the data to be imported we will, therefore we must split the SKU evenly across the customer's branches. In that case the Branch/SKU combination won't be there as it never needed to be, up until now so to do the split, we need the branches created (with no user input beyond doing the import). I hope I've explained myself clearly.
Yes, this is definitely possible. For this, you will want to use the same data source and create some new imports which import into the SKU list to create your new list items. You can put all these imports into a process, then when you run the process and select the imports will run one after another.
For your second part around allocation, I'm sure there are multiple ways to do this and it may depend on what your data looks like, but off the top of my head, I would create a flag which shows the combinations for which you have no historical data and use some formula logic to allocate these evenly.
Sorry, I'm probably not explaining myself clearly. The list items I need to create are for the branch-SKU combinations that need to be created as a result of combining the Customer, Sku forecast data with the list of customer branches, to create a list that's in terms of customer branches and SKUs. If I have a forecast of 30 for Customer Cust1 and product SKU1, where the customer has branch Br1, Br2 and Br3, but no actuals for that SKU1 for any branch of that customer for the same period, then I need to create the list items Br1-SKU1, Br2-SKU1, Br3-SKU1 where each branch has 10 of the original 30. My issue is creating those new combinations; the allocation I'm fine with doing via a simple formula.
Those new Branch-SKU combinations (Br1-SKU1..) don't exist anywhere in the data directly; we have the list of customer branches; BR1, Br2, Br3 and the list of skus; SKU1 but I don't have anything I can import into Anaplan which has Br1-SKU1.
I'm struggling with how to 'combine' the branches with SKUs in order to have something in Anaplan that can be imported into a list to create these new items.
At the Branch level, do you really need the SKU's from the Customer/SKU or just the numbers? I am assuming you want the SKU's under the Branches (hence your post) but are they really needed?
Are those numbers strictly from the "flat" lists or from the hierarchy?
I am assuming the SKU's to Customer are not one to one as the same SKU can roll up to multiple Customers, correct? Same with Branches, correct?
What I am thinking, not knowing the how big your model and workspace are, I am not sure if this will work but to have a "mapping" module by the Branch flat list (no hierarchy) by SKU Flat with a line item that is a boolean. You should be able to get "home" because the parent of Branch is Customer and you can use that Customer code to do a finditem() on the SKU to Customer. If that combination has a value, the boolean in the mapping module should be a True. Based on that true, you can build a view with SKU's rolling up to Branches.
@rob_marshall They do want the individual figures for Sku by branch, so do need the skus under branches, rather than just the figure for all skus under each branch. Those numbers come from the flat lists, not hierarchical lists. That's correct that SKU to customer isn't 1-2-1.
I think I follow your suggestion. I'll give it a go and see how I get on.