What is wrong with:

=SUMPRODUCT(--(Data!E$2:E$3850-DAY(Data!E$2:E$3850)+1=A4),--(Data!F$2:F$3850=B$3))

where A4 houses: 1-Dec-04 and B3 POE-Cite?

Note that A4 contains the first day date of the month/year of interest.

Dominique Feteau wrote:
A collegue of mine figured it out. The reason it was inaccurate was because
I was telling it start at this date and subtract this many days to count
between. For some odd reason, a few values weren't counted. (e.g. there
were 55 "POE-Cite"'s in december, but it only counted 54. So instead of
having the formula calculate the start and end date on its own, I gave it
specific dates. Here's the solution I got:

=SUMPRODUCT(--(E2:E301838322),--(E2:E3018<38352),--(F2:F3018="POE-Cite"))

works like a charm.

thanx for the help.
Niq

"bj" wrote in message
...

I don't see anything wrong with the formula. what is the inaccuracy?

"Dominique Feteau" wrote:


i have a table that has column of dates and a column of names of forms.

heres the sumproduct function i'm using:


=SUMPRODUCT(--(Data!E$2:E$3850=A4-30),--(Data!E$2:E$3850<=A4),--(Data!F$2:F

$3850=B$3))

A4 = 12/31/04 (Date)
B3 = "POE-Cite" (Form)

only problem is that its not very accurate. instead of using "A4-30", is
there a way to use a specific range? i thought using an "AND" statement
would work, but it isnt.

thanks







--

[1] The SumProduct function should implicitly coerce the truth values to
their Excel numeric equivalents.
[2] The lookup functions should have an optional argument for the return
value, defaulting to #N/A in its absence.