Split and Splat doubt

Jordi Vilamajo · 2024-05-06T14:21:11+00:00

There was an error rendering this rich post.

Hi, I have a dataview where Quarters are displayed in columns.

In Q2 only two months are open.
This dataview has a locking condition based on a cube that points the months that are open

I have realized that when I do a input on Total Year the amount is spread on the weight of full Q2 and then this amount is spread on the two open months.

What I would expect was that the Total new amount was spread directly only on the weight of the Open months. Now Q2 is receiving amount based on the weight of the three months although one is closed.

Am I doing something wrong? Is there a way to have the result I was expecting?

Thanks in advance

Find more posts tagged with

Accepted answers

All comments

Shisong

I suppose that is how split and splat works. When you input the data, it will be spread on the weight of the tabular axis. And when you save the data, it will be spread on the most granular level, month in your case.

It's not user friendly (and hope someone can give a better solution), but one possible work around would be to show free cells only while locking.

Tommaso Riva

@Jordi Vilamajo I would take advantage of @Shisong's suggestion with the 'cube cells locked by' option.

Please refer here to our manual to get further explanation and a good example on that.

I hope it helps.

Jordi Vilamajo

Thanks, it seems that that could solve the issue but I get an error message:

I'm including cube 0031 in the Cube Cells locked by, which is a cube that has ones for open months.

Do you know what could be?

Federica Antonelli

Hi @Jordi Vilamajo, you don't need to have cube 0031 in the dataview layout. Just use it for the condition in the "cube cells locked by".

Jordi Vilamajo

Hi, the error was due having a Database Security Profile linked to that cube. I removed it an the error has been corrected. However the result is the same. I do not understand why it does not work; from the documentation you provided it should be possible. Could be due to the fact that Q and months are part of the same hierarchy and not different entities like in the documentation example?

In parallel, I had opened a Support ticket and from there they informed me that the behavior I was pointing it is by design. And if I tried the Data cells locked by I would get the same result, and It happened.

Federica Antonelli

Hi @Jordi Vilamajo good that you raised a ticket on this issue. It is always better in such a case so that the error can be analyzed with all the context info that is fundamental for debugging purposes (cube dimensions, layout details, selections, security…)

To clarify the feature Cube cells locked by. In this situation:

Cube ACT Sales with DE open - in Q2 only 2 months are open, in my example May and June. The locking condition on the ACT sales cube is the following:

if I change the yearly total to 200m - the Q2 total amount is not distributed on the close month, which is April in my example, see below.

Before:

After:

This is a very simple case leveraging the cube cells locked by feature.

Jordi Vilamajo

Hi Federica,

Thanks for the example. Let me explain the situation based on it. When you plug the 200M in the Total, Q2 receives the 34.31% of the variation, which is the weight of Q2 over the Total Open Qs.

For me this is not what any user would expect. If in Q2 we have only two open months, the amount that Q2 should receive is the weight of May+Jun over the Total Open months: 28.14% instead of 34.31%.

It has no sense to receive amount based on 3 months and spread it on two.

The solution is to remove the Q dimension from the tables and leave only months.

But even this solution shows how strange is the way BOARD does this split: it has no sense to have different splits depending on if you show different levels of the time hierarchy.

Imagine the extreme situation where we were displaying data in Halfs of the year. And that we had only one months open on the first half. This month would receive almost the 50% of the increase plugged at total FY.

I think that this would be a good enhancement for new releases. And thanks for your time again.

Federica Antonelli

Thanks @Jordi Vilamajo for the explanation I understand your point and why support gave a "by design" response.