Split merged bars in bar chart

[Anonymous]

Hi, does anyone know how to split these bars up into individual bars?

It seems like Power BI will sum the values of Days (see Excel file, under the column Days) based on the Level. I actually would like to see individual bars, i.e. Francis should have 3 bars of Executive instead of just 1.

Let me know if I'm unclear, and thank you very much in advance!

[Anonymous]

Thank you both for your replies! Much appreciated for taking your time to help solve my case.

Actually the reason why each Employee seems to have multiple rows is because they are different instances. That is, in Week 1, Francis has recruited 3 Executive roles, and for each role he has taken different number of days to recruit them, hence I have placed them in different rows as different instances.

I tried to use "Don't Summarize", but nothing seems to change. I think this case is somewhat similar to this one I found in the Community Forum (https://community.powerbi.com/t5/Desktop/Make-a-column-not-sum-count/td-p/24672). I suppose I could manually divide them up by adding a column to differentiate each row as a unique one, except that I would have a graph that has a very long legend, with the bars very far apart from each other. But the solution as provided by christianfcbmx seems to be quite viable - and I will work on trying that out. 🙂

Thank you very much again to the both of you!

[v-juanli-msft]

Hi @Anonymous 

Is this problem sloved?

If not, please let me know.

Best Regards

Maggie

Community Support Team _ Maggie Li
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

[christianfcbmx · Accepted solution]

Hi, I dont understand what you are trying to do...there must be other better way... you are trying to say that francis has other dimension that is not declare in your table to say that Power bi has to split in 3... the question is why Francis in the same week under the same level (executive) has 3 differents ammounts of days?... if you dont differenciate that you will always have 1 bar....thats the logic.

1st// solution is to use the days column as x axis as well (categorical x axis not lineal)

2nd// solution is to differentiate the level column like Executive 1; executive 2 and so on.

[Anonymous]

It looks like this graph is automatically summarizing your like data.  So all of James' time is executive time, so he gets 1 bar of sum'd executive time.  Linda gets 3 different bars because all of her time is different somehow.  Francis has 1 bar for all his week 1 executive time because it's all like data.

On some visuals there's an option for how to treat your data called "Do not summarize", which would stop it from doing this.  However, it doesn't seem to be available on this graph. 

The explanation I found for that was, more or less, it did not make sense not to summarize data in this kind of visual, and that in this case your best bet may be to transform your data somehow (and I'm looking into how you might do that).

EDIT: I didn't have much luck transforming it meaningfully, but maybe a more experienced user will have a solution.