Hi, put a default on your value in the place you are presenting it; @{P_Year}{WhatEverYouWantHere} then the user will see the default when it is not populated

Wow! Anyone with psychic powers please raise my hand? Yes, thanks SonPat99 that is what I was suggesting, if that is what he wants...

Now you lost me, are you saying you want; - A. SUM(MAX(fact column by dim1) BY dim2, dim3)/SUM(MAX(fact column by dim1, dim2, dim3) BY dim3, dim2) or B. avg(SUM(MAX(fact column by dim1) BY dim2, dim3)/SUM(MAX(fact column by dim1, dim2, dim3) BY dim3, dim2)) ??

Did you get there in the end? If so can you kindly close the question / mark correct / mark helpful.

Is this with the average logic in the formula? avg( whatever you had before) ?

I know it is not an ideal solution, but I recommended the use of 'server complex aggregate' because it pushes the calculation back to the database, aggregation and all. Like you standing on the high ground of good architectural practise I prefer measures in the rpd, but from past experience the sum by two dimensions…

On your measures that are coming out incorrectly have you tried ticking 'server complex aggregate' for those measures?

Simple answer, whenever you are looking at division the measure must be physical not logical. So recreate a physical measure based on the physical not logical columns - it is down to the difference between aggregate first then divide OR divide first then aggregate.

No problem! Glad of your input!

This was my suspicion when the OP was saying that the function result was 'identical'. I was once pulling my hair out with something that was identical, only to discover it was padded with 6 nulls (seriously) at the end of the string.

This was the previously;- "Okay, try this. FILTER("Historical Pipeline Facts"."Open Opportunity Revenue" USING ("Pipeline Snapshot Date"."Pipeline Snapshot Enterprise Quarter" = trim(CASE WHEN RIGHT('2018 Q 2', 1) = '1' THEN CAST(CAST(LEFT('2018 Q 2', 4) AS INTEGER) -1 AS CHAR(4))||' '||'Q'||' '||'4' ELSE…

Yes, I posted that previously?

Csv?? What csv?? It looks like you just put a substring around the year part of the formula result then, must be something weird in casting an integer to a char(4).... Then it should be job done.

Okay, strange agreed, can you find out what the ascii of a / the characters after the 8th is?

I need closure, did you check, are both exactly the same?

Is the answer to both; - 8??

Can you put one of the formulas through length and see if it differs from if you put your quarter through length?

Okay, try this. FILTER("Historical Pipeline Facts"."Open Opportunity Revenue" USING ("Pipeline Snapshot Date"."Pipeline Snapshot Enterprise Quarter" = trim(CASE WHEN RIGHT('2018 Q 2', 1) = '1' THEN CAST(CAST(LEFT('2018 Q 2', 4) AS INTEGER) -1 AS CHAR(4))||' '||'Q'||' '||'4' ELSE CAST(CAST(LEFT('2018 Q 2', 4) AS INTEGER) AS…

So try this; - FILTER("Historical Pipeline Facts"."Open Opportunity Revenue" USING ("Pipeline Snapshot Date"."Pipeline Snapshot Enterprise Quarter" = trim(CASE WHEN RIGHT('2018 Q 2', 1) = '1' THEN CAST(CAST(LEFT('2018 Q 2', 4) AS INTEGER) -1 AS CHAR)||' '||'Q'||' '||'4' ELSE CAST(CAST(LEFT('2018 Q 2', 4) AS INTEGER) AS…

Okay, do a replace on the above just to prove it is catching the prompt please; - FILTER("Historical Pipeline Facts"."Open Opportunity Revenue" USING ("Pipeline Snapshot Date"."Pipeline Snapshot Enterprise Quarter" = (CASE WHEN RIGHT('2018 Q 2', 1) = '1' THEN CAST(CAST(LEFT('2018 Q 2', 4) AS INTEGER) -1 AS CHAR)||'…