Skip to main content
cancel
Showing results for 
Search instead for 
Did you mean: 
samuel_lin
Creator
Creator

Table Transformation - Customer Journey (Products)

Hi Qlikers,

If I have a table:

CustomerIDOrderRankPurchased Product
1231productA
...1productB
...2productA
...2productK
1233productJ

and I would like to get the following table:

CustomerIDFirst Purchased ProductSecond Purchased ProductThird Purchased Product
123productAproductAproductJ
...productAproductKproductJ
...productBproductAproductJ
...productBproductKproductJ

so then in aggregate, i can look at the number of customers who purchased from productA to productA to productJ, basically to figure out the most popular customer purchasing path.

Does anyone know an easy way to achieve this table transformation?

Thank you!

Samuel

2 Replies
bwisealiahmad
Partner - Specialist
Partner - Specialist

Hi,

Hope this helps. You could actually maybe do it even more more smooth with a For Each Loop, but I'm off to bed so hopefully this helps:

Put this in your script:

Temp_Numbers:

LOAD * INLINE [

    CustomerID, OrderRank, PurchasedProduct

    123, 1, ProductA

    123, 1, ProductB

    123, 2, ProductA

    123, 2, ProductK

    123, 3, ProductJ

];

Facts:

// First Products //

NoConcatenate

LOAD

CustomerID,

PurchasedProduct AS FirstPurchasedProduct

Resident Temp_Numbers

WHERE OrderRank = '1';

// Second Products //

Left Join (Facts)

LOAD

CustomerID,

PurchasedProduct AS SecondPurchasedProduct


Resident Temp_Numbers

WHERE OrderRank = '2';

// Third Products //


Left Join (Facts)


LOAD

CustomerID,

PurchasedProduct AS ThirdPurchasedProduct

Resident Temp_Numbers

WHERE OrderRank = '3';


Drop Table Temp_Numbers;


That gives you this:

Capture.PNG

Best,

Ali A

rwunderlich
Partner Ambassador/MVP
Partner Ambassador/MVP

I think you can do it with a single Generic load statement:

Generic

LOAD CustomerID,

    'Purchased Product #' & OrderRank,

    [Purchased Product]

FROM ....

See Use cases for Generic Load | Qlikview Cookbook for a similar example.

-Rob

http://masterssummit.com

http://qlikviewcookbook.com