One way to get around duplicates in this type of setup
set up a helper column
if your comparison data is in column A
enter in helper column
=rank(A1,A:A,1)+row()/100000
this will give a unique number for matches
Note that the higher row numbers will be ranked higher in this case.
if you want lower row numbers to be selected first, use a minus row...
If you have an alternate tie breaker it can also be used
(I have had complicated enough tie breakers that I had to go to 6 layers of
rank to stay within the 15 digit limit for Excel.

"Richard" wrote:

I am using INDEX, MATCH & LARGE function to find a customer reference for top
3 values in a specific column in an array of data;

Basically I use LARGE (1...3) to find the top five values, and use the MATCH
function to find the row reference in the array.

INDEX function then points to the array Row and Col for the customer
representing that value.

My problem is that the Match function always finds the first value if there
are multiple equal values in the column. So if 2 values are equal, then
match always picks the first of these values, even though Large 1 & Large 2
will find both values. How do I create a pointer reference to the second
value (as selected by the Large function)

CU_Array = $A$1:$C$5
REF = 1,2,3 can be one of 3 values
A B C
1 Cust_Row1 100 200
2 Cust_Row2 67 150
3 Cust_Row3 100 125
4 Cust_Row4 156 200
5 Cust_Row5 120 100

The formula I am using is
=INDEX(CU_Array,MATCH(LARGE(c1:c5,REF),c1:c5,0),0)

Anyone have any suggestions...

Thanks Richard