Hi slorin 

Your solution works really well!

My powerqeury became so fast that I even have some time left to ask some other stupid questions... 🙂

What I really don't understand is the difference in performance between

1. parcels-manual

just by using the GUI, combining 2 columns "Capakey" and "Partnumber": 

let
    Source = Excel.CurrentWorkbook(){[Name="Table18"]}[Content],
    #"Merged Columns" = Table.CombineColumns(Source,{"Capakey", "Partnumber"},Combiner.CombineTextByDelimiter(" | ", QuoteStyle.None),"PAND")
in
    #"Merged Columns"

2. parcels-fixed

First, I defined a list "DATADEF-PAND-fixed"

let
    #"Result" = {"Capakey", "Partnumber"}
in
    #"Result"

that I use in my query "parcels-fixed"

let
    Source = Excel.CurrentWorkbook(){[Name="Table18"]}[Content],
    #"Merged Columns" = Table.CombineColumns(Source,#"DATADEF-PAND-fixed",Combiner.CombineTextByDelimiter(" | ", QuoteStyle.None),"PAND")
in
    #"Merged Columns"

Do I need to conclude that calling any function (i.e. "#DATADEF-PAND-fixed") always causes a delay, even when this function doesn't do anything?

And I this function in the above example excecuted for every record in my dataset? Or why does it cause so much delay?

Is there any drawback on buffering? Or should I always buffer lists? And if so, why is this not the standard behaviour?

The documentation is very sparse on this
List.Buffer - PowerQuery M | Microsoft Learn

The documentation for Table.buffer is a little bit more elaborated

Table.Buffer - PowerQuery M | Microsoft Learn

Here I can read

In some cases, it can make your queries run more slowly due to the added cost of reading all the data and storing it in memory

So I'm a bit confused when and when not to use the buffer....

kind regards

bartplessers