My question pertains to the MB Affinity tool, which is simple and powerful, so far I'm liking it a lot. My question is about reporting this data and whether or not it's possible to export the data in a Tabular format.
Take an example with 5 items, the MB affinity tool will produce a 5x5 grid with 25 values, all representing coaffinity scores. I currently write this table to a SQL Server where I perform an Unpivot function similar to the below.
select u.rownames, u.Segment2, u.Affinity
from MB_STG_families A
unpivot
(
Affinity
for Segment2 in (Yuengling_Traditional_Lager,Wraps,Whiskey,Vodka,Unknown_Segment)
) u;
This gives me an output like the below
rownames Segment2 Affinity
AF___Lemonade Yuengling_Traditional_Lager 0
AF___Lemonade Wraps 0.0760472674700328
AF___Lemonade Whiskey 0.00555589852284023
AF___Lemonade Vodka 0.0209656967344384
AF___Lemonade Unknown_Segment 0.029074190231479
In my case I am then running this MB affinity once each for our 5 principle customer segments and comparing affinities across segments. Without unpivoting the data I believe this analysis would be difficult.
My question is: Is there an ability within Alteryx to report this data in Tabular form from the get-go? We sometimes perform the analysis on a larger set of items (in my example there are 5, in reality there are hundreds/thousands) and so this becomes more complicated. I know that I can write a SQL query to identify column names and automatically perform the Unpivot, but I believe having the data in Tabular is more useful anyway, and am hoping Alteryx has this in mind.
My second question is: The coaffinity scores seem direction-agnostic, is there a vision to change this to "When Item 1 is purchased Item 2 is purchased at X score" AND "When Item 2 is purchased Item 1 is purchased at this score". I could be misunderstanding the score, but I believe it's not capturing that nuance.
Thanks
