I have a simple list containing values, but some of them have the same values. So when I rank them, the ranking is quite discontinued. For example, 1,1,3,3,3,6,6,6,6,6,6,6..... but what I want is 1,1,2,2,2,3,3,3,3,3,3,3,....... How to approach it.
Here's one way, which is probably the most expeditious and reliable, also one that can be automated but it does require you to run an action.
We'll create the following:
Transaction Module (holds your 1 dimensional module and the values you want to rank)
System Module for the transactions
A List that contains all the unique values of the value you want to rank
An action that creates the list above
A system module for the list
First, Create the transaction module. You mentioned no dimension so I assumed you meant its a transaction module with 1 dimension. In this case I chose an arbitrary unique identifier using a sequential number.
Note: you can see the last line item provides the ranking you want!
Secondly we create a system module for the transactions. This follows the DISCO methodology - calculate once refer often. Make sure you create a saved view that only contains the list items. There should be no aggregate values in the list or you will get errors in the next step.
Next, we create the List of unique values using the system module we created above. Create an import action to do this and save/rename your action because you will need to run this action everytime you want to reroll the rank. You can add this action to a dashboard to make it easy to access.
Now that we have a list of unique values we can rank them in our system module for the list you created above. Again, use the DISCO methodology and set all summaries to "none".
As you can see the "Rank" line item contains the right value we want. Last step is to add a line item to our transaction module (or system module) that contains the final rank.