I've got following table:

Invoice Contract Amount Client Assignement
#N/D GIFE 0051/08
#N/D GIFE 0091/08
1008-08 2008.30.3009 555.00 BMG 1008-08
1008-08 2008.30.3009 555.00 BMG 1009-08
#N/D GIFE 104/08
116-08 2008.30.3009 606.40 BMG 116-08
#N/D GIFE 119/08
151-08 2008.30.3011 606.40 BMG 151-08
#N/D GIFE 164/08

In this table, Assignements are unique, but Invoices aren't.
I'd like to summarize my data so that I get the sum of unique
Invoices; as you can see in row four, Assignements 1008-08 and 1009-08
are both covered under Invoice 1008-08, for an Amount of 555.00; I've
paid 555.00, not 1110.00; I entered it twice to have in any row the
whole information.

If I try to build a pivot table, I get for Invoice 1008-08 a sum of
1100.00, which is the standard behaviour: is there a way to sum only
unique values in a pivot table?
How could I solve this question otherwise?

Please tell me if in your opinion this is only a problem in the layout
of my table... e.g. I could leave blank any second, third, fourth
"Amount" of the same "Invoice" (which is what I've done now, waiting
for more elegant solutions).
Ciao,
Dan