Pareto analysis with classification. How filter a single customer (whit correct classification)?

[#1]

Hi All,

I created a Pareto analysis. It's correct.

But when I filter a BP (customer) the classification ('A', 'B', 'C')  it's no longer correct (sales become 100% ).

How can I visualize the correct classification filtering a single BP?

Thanks

Daniele

[hic]

You need to replace Sum(T_Ordinato) with Sum({1} T_Ordinato) in a number of places, including the sort expression. In the picture below, I have done so in all places except the first expression. This way you can see the selection there.

HIC

[jagan]

Hi,

Check this expression for Prog. Sales % €

RangeSum(Above(Sum(T_Ordinato), 0, RowNo())) / Sum({<T_BP=>} total T_Ordinato)

Regards,

Jagan.

[#4]

Hi,

with this expression, the classification "A" it's not correct for DI2 customer. It should be "B".

For Henric: I would only 1 row (this report will send to BP)

Thanks

[hic]

If you want the classification to be static, you should probably do the above calculations already in the script. I.e. you need to create a table with the total sales number, grouped by customer, sort it according to sales, and assign the classification. Then you will have the classification static and in a field of its own.

HIC

[jagan]

Hi,

If you want to retain the same values then you have to calculate this in script.

Regards,

Jagan.

[#7]

Hi Henric and Jagan,

the problem is that this classification is dynamically determined (for Company,Year, to Month...) during analysis

I can't do this in script.

Perhaps the problem has no solution...

Even Qlik has limits ... 

Regards

Daniele

[#8]

Yes,

Actually in this scenario, your calculations are being calculated on the fly, functions like rowno() and Above are chart level functions and when you make selection, these functions again change the values according to your selection,thus you get results on the fly, this is the good part of Qlik 🙂

[hic]

An optional way to do this is to use an ABC-categorization based on rank rather than Pareto. Then you can define a calculated dimension where you explicitly disregard a selection in the Customer field.

Aggr( If((Rank(Sum({$<Customer=>} Sales),1)-1) / Count({$<Customer=>} distinct total Customer)< 0.50, 'A',

          If((Rank(Sum({$<Customer=>} Sales),1)-1) / Count({$<Customer=>} distinct total Customer)< 0.75, 'B', 'C')),

          Customer)

See more on Recipe for an ABC Analysis

HIC