Measures as a dimension?

[AlexaderMilland]

So I don't know why I can't get the grasp on how to do this.

For a simple version i have an "Orders" table with just 3 columns

OrderID | CustomerID | Date 

I want a matrix visual that shows me 

How many customers purchased a total of 1 order, 2 orders, 3 orders etc etc. 

I simple can't find a way to do this using measures to distinctcount orders and customers respectively.

I need it to be measures, since the user input has to be a date range or similar (or other filters in the model which is vastly greater than the simplified table above). 

Output would simple just be something like: 

#Orders | #Customers

1 | 200

2 | 75 

3 | 47 

4 | 35

5 | 21

I can easily do this as a calculated column on a seperate customer table, but that wont allow dynamic filtering/slicing.

[Greg_Deckler]

@AlexaderMilland Create a separate, disconnected table to store your order numbers in a single column

1

2

3

4

5

Put that in a table visual along with this measure:

Measure =
  VAR __Count = MAX('Disconnected Table'[Orders])
  VAR __Table = SUMMARIZE( 'Orders', [CustomerID], "__Value", COUNTROWS('Orders') )
  VAR __Result = COUNTROWS( FILTER( __Table, [__Value] = __Count ) )
RETURN
  __Result

[AlexaderMilland]

You mean an index table that just goes from 1 to infinity (or well some max i set) ?