Filter aggregation visual by ID field

[victor_erathos]

My question revolves around two tables, one with the information of each distinct product and another with the amount of users associated with each product in different cities, such as the following example:

My goal is to display the "number" field of products aggregated by category, but be able to filter which BARCODES (i.e. products) should be considered in the aggregation based on the number of users. I set 3 ranges of users, which I want to use to filter the products with this disconnected table:

The main problem is that the number of users is not fixed for each product, but rather by product and city. That is, I need to able to filter which cities should be considered when accounting for this. I created this measure to serve as a flag if the SUM of users is inside the selected range(s).

flag =
IF (
    AND (
        SUM ( Users[users] ) >= MIN ( pRange[lower_limit] ),
        SUM ( Users[users] ) <= MAX ( pRange[upper_limit] )
    ),
    1,
    0
)

The following screenshots illustrate what my expected output is with the help of a table visual on the side. Both visuals have the "flag" calculation added as a filter to them (filter: is 1).

This first image displays all the data, with all cities selected and all possible ranges.

When I try to select a city and a range, the table displays exactly what I expect, only the products that satisfy the condition. However, the bar chart naturally tries to aggregate all products that fall in each of the three categories and then evaluates the number of users, which in this case doesn't fall in the "Low" range, thus, the chart is blank. However, what I want is for the bar chart to categorize exactly the products that we see on the table, which is essentially filtering the chart by barcode, based on the condition.

I attempted to create another table that would do exactly this, to then be able to use as a filter. This is one of the "SUMMARIZE" structures that I tried to create a reduced table with only the cities selected, which didn't work so I didn't add the range filter which is also necessary as well (please note that multiple cities or ranges can also be selected).

attempt =
SUMMARIZE (
    FILTER ( Users, Users[city] IN ALLSELECTED ( pCities[city] ) ),
    Products[barcode],
    "users", SUM ( Users[users] )
)

All example data as well as this "attempt" are available here in this .pbix file.

Thank you in advance for the help! I'd be happy to further explain the problem if necessary.

[Anonymous]

Hi  @victor_erathos ,

Here are the steps you can follow：

1. Create measure.

User_city =
var _selectcity=SELECTEDVALUE('Users'[city])
var _selectrange=SELECTEDVALUE('pRange'[range])
var _usercity=CALCULATE(SUM('Users'[users]),FILTER(ALL('Users'),'Users'[city]=_selectcity&&'Users'[barcode]=MAX('Products'[barcode])))
return
IF(
_usercity =BLANK(),0,1)

Min =
var _selectrange=SELECTEDVALUE('pRange'[range])
var _min=CALCULATE(MIN('pRange'[lower_limit]),FILTER(ALL(pRange),'pRange'[range]=_selectrange))
return
_min

Max =
var _selectrange=SELECTEDVALUE('pRange'[range])
var _max=CALCULATE(Min('pRange'[upper_limit]),FILTER(ALL(pRange),'pRange'[range]=_selectrange))
return
_max

Flag111 =
IF(
    [User_city] >=[Min]&&[User_city]<=[Max],1,0)

2. Put [Flag111] in Filter and set is=1

3. Result:

Please click here for the pbix file

Best Regards,

Liu Yang

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly

[victor_erathos]

I am not sure I understand your approach to the problem. It still does not seem that I can create a bar chart as I mentioned in the post — agregated by category with only the relevant products (please also bear in mind that I created a simplified example, in my actual data I have more than 100k unique products).