How to separate different kinds of differences

[datanibbler]

Hi,

I am just comparing two different possibilities to do essentially the same:

- Right now, we are using Crystal to generate our turnover which is suboptimal in several points (I won't go into details here, I don't
   want to risk my prejudices, I have worked hard on them 😉

- I want to replace that (or those because there is not one Crystal_report for turnover, but we have to use several) by one
   QlikView_report.

To that end, I am reading from an Excel file that has been generated out of the CR for June and at the same time I'm running the QlikView_app for June.

- Now I have a nr. of records in the CR and one in QlikView. There are approx. 3.000 items'  difference.

- However, when I

     - select all selectable items in the listbox of the items_in_QlikView and then

     - select all excluded from the listbox of the items_in_CR
       (both tables are linked, but the fields I do the count on are not the keyfield)

     => I get a count of approx. 6.000 items' difference.

=> So I guess that the sheer number of items differs by about 3.000 while the other 3.000 just don't match for any reason.

=> How could I separate these?

Thanks a lot!

Best regards,

DataNibbler

[datanibbler]

No, I think that was a thinking error,

- the 2.462 (difference in the sheer nr. of records between CR and QV) are contained in the 6.612 (items in CR without a match in QV).

So I need to try to eliminate the non-matching items bit by bit and eventually I will have only a nr. of matching items - and a nr. of items that are just not present in QlikView.

I will try. I think approx. 1/3 (about 2.000) may be due to the CR "looking at" a different date than QlikView, so that the CR shows a number of items that fall outside the timeframe that I have selected in qlikView.