Gains/Pareto chart

[Anonymous]

Hi, 

I'm trying to make a graph, where I can express what percentage of items sold equals to what cumulative sales amount. E.g., 50% of the total sales are represented by x% of items. Below is an example chart:

A small example of my dataset can be seen below: 

Sales Amount	Item	Date
412	40000137	2019-06-18 00:00:00.000
850	40000137	2019-06-19 00:00:00.000
883	40000137	2019-06-20 00:00:00.000
650	40000137	2019-06-21 00:00:00.000
13	40000153	2019-06-18 00:00:00.000
44	40000153	2019-06-19 00:00:00.000
122	40000153	2019-06-20 00:00:00.000
88	40000153	2019-06-21 00:00:00.000
1.182	40000167	2019-06-18 00:00:00.000
5.065	40000167	2019-06-19 00:00:00.000
4.665	40000167	2019-06-20 00:00:00.000
4.743	40000167	2019-06-21 00:00:00.000

I have around 4000 items for the past 2 months, så I'd like to be able to filter on dates as well. How can this be achieved? 

Thanks in advance

[vanessafvg]

@Anonymous  ive sort of worked out the basics for you, see if this works.

[Anonymous]

@vanessafvg  Thank you, but I actually managed to get that far by myself. 

I'm struggling with finding out, how to have both axis be percentages. Essentially, I'm trying to get a coordinate set of percentages: 
%cumulative_sales, %cumulative_items and be able to modify/filter the date range for this calculation. Using measures only gets mere halfway there. 

[vanessafvg]

@Anonymous  i am not sure i actually understand what you are asking.  please show me exactly what you are looking for if you convert the graph to a scatter plot you will have both cordinates plotted.  If you put the date in a slider it will shift according to the date?

can you please identify what exactly your issue is, is it a visual that you are looking for, you say i you have the %'s arleady, it would have been better for you to give me everything you have done so far so that i didn't duplicate what had been done already.

[Anonymous]

@vanessafvg 

The calculations you made are based on the dates. The %cumulative sales and %cumulative items are calculated correctly (I think), but the problem is to plot them as "what part of the sale is caused by what part of the items". 

As an example from the data I originally posted.

Item	Total sale%	Item %	%cumulative Sale	%cumulative item
40000167	83,64%	33,33%	83,64%	33,33%
40000153	1,43%	33,33%	85,07%	66,67%
40000137	14,93%	33,33%	100,00%	100,00%

Here %cumulative. item should be on the x-axis and %cumulative. sale is the y. Graphing this would yield a somewhat concave graph, which allows to look at what share of the sale is explained by some share of the items. 

Furthermore, I should be able to filter on the dates, so that the calculations are based on e.g. 3 of the above days and not all 4. Changing the date range, should change the calculations. 

Also, I already posted the type of visual/result I am looking for in the OG post. 

[vanessafvg · Accepted solution]

@Anonymous  i am afraid the visual you are asking for does not exist in power bi.  Maybe you can find one in the custom visuals but i couldn't see anything

i have made a few adjustments which maybe you can see a bit better what you were asking for.    Because there are the same amount of items for everyday you wont see any adjustment between the 33% split between all of them per day but the sales adjust tallying up to 100% depending on what day/s you have selected

[vanessafvg]

@Anonymous  please provide your data in text format

[Anonymous]

@vanessafvg done