Are there any plans to allow more than 3 dimensions in a pivot, I have a situation where that would be required. Or are there any work arounds?
You can post the idea on Anaplan idea exchange forum and talk about It’s benefits and use cases.
@alismith here's an idea (workaround) you can try. It's a lot of steps but you can automate the whole thing with actions.
I use this when I want to export data flat with no sparsity.
I second @Misbah recommendation to add to the idea exchange!
- Four dimensions Product, Location, Customers, Reps
- We want to see all dimensions on the rows
- Create a single list using the 4 dimensions as a key
- Dimensions become Line Items
1. Using the DISCO methodology make sure you have system modules for each list.
See @DavidSmith post:
2. Next Create a calculation module that gathers all the data and calculates your Key.
Note: it's okay that the calculation module only shows 3 dimensions. We'll fix that in a bit.
Note: you can also add a Boolean to filter only the records you want, thus reducing the size of the Output module in the next step.
3. Import the key and create a list. To make this reusable, you can delete the list first, then repopulate it with actions.
4. Finally, create the output module and display all the dimensions, and if you used a boolean, only the data you wanted will show up.
Note: You can accomplish this with formulas or by using an import. I took the shortcut and used an import.
Note: This is at the grain. If you want to subtotal, you can create another module that sums on the list total or parent if you use a structured list.
Note: I took the key to 6px since it doesn't have much meaning on a dashboard
Note: you could create a concatenation of two of the dimensions instead of all four to keep the dimensions to 3. Then you can use the summary logic (subtotals)
hi @JaredDolich , in the output module are you using all 4 dimensions again?
or is the new list the only dimension?
And if it is the only dimension how are you linking the 1 dimensional model to the 4D model? Any lookup i try to use seems to break
For the sake of discussion, assuming Anaplan did increase the number of dimensions you can nest along the row or columns, the question i would pose to you is: should you as the model builder/solution architect suggest/build such a module, i.e. a module using > 5 lists (this is technically a separate but relevant issue).
Please remember the first guideline in Planual, i.e.
Besides the most obvious implication of increasing workspace exponentially, it also has the following downside to building a huge module:
1. It's difficult for user to select the right combination of list member to arrive at the desired combination,
2. It's difficult for model builder to write formulas that cater for all scenarios and list combinations.
3. Performance would suffer because you're likely to turn on Summary for some of the line items (i'm imagining you're requesting for a multi nested row/column because this module is an Output type module, and end users likely want to see aggregation of some sort).
Having said the above, as mentioned by @Misbah this >3 nested row/column option is not available at the moment, so please upvote the relevant ideas exchange post.
However i would suggest that you do 1 of the following:
1. Refer to post by @JaredDolich
2. Break up the single module into multiple module and publish them in the same dashboard, followed by user education.
3. Export the data into another analytics tool.
The output module can be any number of dimensions but in this case I combined two of them so I could nest them on the row pivot, essentially giving me the effectiveness of four dimensions but only having to use 3 (max).
Make sure to read @LipChean_Soh commentary on best practices (above) and @Misbah post as well. These are important points I didn't mention. To be honest, I kind of like the 3 dimension limit because it forces me to break up the process into manageable pieces and helps with user adoption. Unlike Excel where you have unlimited space to work with, Anaplan does force you to use a narrow amount of real estate so the steps have to be simple.
Then again, sometimes, you just need all those dimensions. We all face that challenge.
Great question and discussion. I hope you'll let us know what you decided to do.