About usman.zia

Now solved!    I created a module dimensioned by the 9 line items as a list and referenced the 9 line items to the list from the source module and used first occurrence along the 9 line items dimension to get the number of unique names.    Thanks   Usman ... View more

Hi @jonathanpascual    As @andre.lie mentions you need to change the summary method to allow the formula to show the value at the higher level.    But you need to consider a few points: 1. Does your hierarchy have multiple items below and is bringing the last or first the correct code you want to see? (as in @andre.lie example he has many cities and if they all had a code which code would you want to see?) 2. It may be the case that you only have one item today but in the future this may change you may have two items and this can cause errors in the future (has happened to me!). How do you deal with this? 3. Do I need to have another formula at the top level to account for this and change the formula to bring this in to avoid the issue (where a subsidary view can help).   Also I notice in your formula you have hard coded codes is this just for the result of the example your explaining or it is actually hard coded? Worth considering is this sustainable for the future?   I hope this helps!   Thanks,   Usman ... View more

Thanks for sharing @Misbah    Great tips @LipChean_Soh    For me it is clear that fake versions are much more flexible than native versions. There are so many product limitations when using native versions as you mention: - Cannot use line item format as version - Cannot use LOOKUP - Cannot use SUM - Cannot use subsets of versions I believe versions with the above functionality has been expressed on the idea exchange and I can't wait till it is on the roadmap!  Therefore in most cases I'll will always begin using fake versions due to flexibility.   The specific case that you mention which requires a change is only as complex as the model you are trying to implement the change of native versions to fake versions in.  If it is a model with versions in most places with great complexity then this will take time and if the model is big in size taking up a large part of the workspace it may be difficult.    The best way to approach this is to first create a fake version list with only a small number of versions such as actual and forecast. Create module shells of the modules you wish to copy (module shell = module with line items but no formulae and minimal dimensions (consider using subsets)) these will be all the modules with native versions. Easiest way is to copy existing modules and then delete formulae and delete versions etc. The risk you have to consider here is that the data stored against the native versions will most likely be lost so it is worth creating exports and saving the data so you can re-import it at a later stage. Check your actions and imports! you will need to note the ones you need to recreate because of the changes and in some case you will need to delete them as they may prevent you from making changes. Once you have completed all the shells, export the formulae from the modules with native versions and delete the native version modules. Evaluate the formulae and begin pasting them back, expect some dimensionality issues it won't completely  straight forward. Re-create your imports and you should be good to go! But make sure you TEST!    If you had unlimited space you could copy the model and do a similar task but you wouldn't have to delete the existing native version module right until the end because you want to compare and export the data from those modules.    I hope this gives some guidance!  It will be challenging but it can be done!    Thanks,   Usman ... View more