Hi all - I've received some terrific info here before, and after 2 days
of struggling have decided to try again. Here's my problem:

I have a PivotTable tied to an ODBC data source (AS400 Table). The
data is updated daily. A simplified version of the table is:

A B C D E
1 Sales Acct Dr Item Catg
2 Bob 100 Dr. A Zig Res
3 Bob 100 Dr. A Zag Res
4 John 101 Dr. B Zig Res
5 Tim 121 Dr. C Zog Ox
6 Tim 121 Dr. C Zag Res
7 Bob 133 Dr. D Zog Ox
8 Tim 141 Dr. E Zig Res
9 John 167 Dr. G Zog Ox
10 John 167 Dr. G Zag Res


My PivotTable appears as follows:

Sales Dr Acct Catg Item Total
Bob 3
Dr. A 2
100 2
Res 2
Zig 1
Zag 1
Dr. D 1
133 1
Ox 1
Zog 1

John 3
Dr. B 1
101 1

Res 1
Zig 1
Dr. G 2
167 2
Ox 1
Zog 1
Res 1
Zag 1

(I hope that doesn't look to confusing).

What I need is to calculate - based on the PivotTable data - how many
of each category (Res, Ox) each sales person (Bob, John) has sold. I
also need to exclude items that aren't tracked (Zig's). The sales
manager viewing the spreadsheet needs to see all of the sales,
regardless of whether the item is being tracked. So, in the above
example Bob would have 1 "Res" sale and 1 "Ox" sale. John would also
have1 "Res" sale and 1 "Ox" sale, because "Zigs" aren't tracked. If
this makes sense to anyone, I'd appreciate any input you could give.
Thanks in advance.