I think this one warrants a quick macro, give this a try (on a copy of your data in case of problems! I have tested this on Excel 2003 here, and it works for me, but as always it's best to be cautious!).
First, it will select the entire sheet you currently have active and sort by the
A column. Then it will then look down the entire
A column for things than match (100% match, this is case sensitive as well) and will add up thier values in the
B column and remove the duplicate rows. Data in the duplicate rows in columns other than
B will be lost.
I've added a couple of
NOTE comments in the code with hints on the bits that are easiest to tweak.
'NOTE: Change this select if you wish the sort to be more precise
Selection.Sort Key1:=Range("A1"), Order1:=xlAscending, Header:=xlGuess, _
OrderCustom:=1, MatchCase:=False, Orientation:=xlTopToBottom, _
'And then merge
'Keep going until we run out of entires in the first column
Do While ActiveCell.Value <> 0
'Loop while the row below matches
Do While ActiveCell.Offset(1, 0).Value = ActiveCell.Value
'The value on this row += the value on the next row
'NOTE: Changing the 1 in the second places on *all three* of these
' offsets will change the row being merged (A+1=B, A+2=C, etc)
ActiveCell.Offset(0, 1).Value = ActiveCell.Offset(0, 1).Value _
+ ActiveCell.Offset(1, 1).Value
'Delete the duplicate row
'Reselect the top row for this group
'Step to next row