Actually you gain very little by splitting one table into many. The way Qlik Sense works with its associative database and de-duplication (compression) it will be more efficient and speedier in keeping it in one table. So why do you need to split it in the first place?
it's not the load of the table itself, that is bothering me.
Later on in the script I do some calculations which are different dependent on the product and each record can contain info on each of the products. So I have to:
For every record check if it contains product A data; then do this
For every record check if it contains product B data; then do that
and so on
I think I already gained some speed because I decided to split the table and have to do the checking only once instead of several times (and by reducing the amount rows that I have to read not even containing any relevant data).
I was just wondering if I could gain some more speed, by being able to have the split in one go. But I see the point that a simple LOAD cannot provide this.
actually I am turning the data into a cross table further down the road to be able to do other things in the reporting itself. It's what I do (have to do) with the data before I can convert it to a cross table, that's bothering me and makes me wonder if a cross table what help at that point. But it's worth some additional thinking. I'll keep you updated, if it helped.