Skip to main content
cancel
Showing results for 
Search instead for 
Did you mean: 
bmachado7
Contributor
Contributor

Sum with multiple conditions

Hi everyone,

I'm a new qlik sense user and I've been racking my brain for hours without getting anywhere.

I have a table with sales splitted by product and cities. 

I need to sum sales of all cities where Cluster Special' Products were sold (in the example, I have Product A in special cluster)

So, in the table below, product A was sold in city1 and city3. I need to understand the percentage of total sales these 2 cities represents, regardless of the product.

Once identified the cities , I will sum all sales (regardless of the product or cluster) of them (city1 +city3) and then divide it by total sales to find out how representative they are. 

The problem is that I'm not being able to create a condition that sums the sales of the cities selected regardless of the product. It always considers only sales of special products in the cities. I need it all. Cluster field should only help me to identify which cities I should sum.

One of my many attempts of calculation was:

If(CLUSTER='Special' and Sales>0,
sum({<[ANO]={$(=Max([ANO]))},>}$(=[Sales]))
,0)

CityProductClusterSales
City1ASpecial75
City2ASpecial0
City3ASpecial41
City1BOther23
City2BOther97
City3BOther86
City1COther99
City2COther84
City3COther68
  Total sales573
  City1+City3392
  %represent68,4%

 

Can someone help me, please?

3 Replies
Channa
Specialist III
Specialist III

If(CLUSTER='Special' and Sales>0,
sum({<[ANO]={$(=Max([ANO]))},>}[Sales])
)

/

 

sum({<[ANO]={$(=Max([ANO]))},>}[Sales])

 

Channa
bmachado7
Contributor
Contributor
Author

Tks, Channa

But this syntax only considers sales from cluster special. I was hoping to sum all sales of all products of cities that have sales of special cluster

The cluster information should be used only yo id the cities I have to sum. The sum itself should consider all products sold in the city

marcus_sommer

I think it's only with an indirect set analysis lie p() and e() possible to set the conditions. Take this as a starting point for your task:

sales.JPG

- Marcus