Hello MisterExcellent and Marcelo,
I was looking at your last formula, Marcelo - I don't think you need CSE, by the way, the results should be the same without it - but I did find some discrepancies, e.g. if B5 is 26/05/2007 and C5 is 30/04/2008 then the result returned is 01/05/2008 - 26/08/2007
Also I think there are oddities with dates at the end of the month, e.g.if B5 is 30/05/2003 and C5 is 25/11/2005 then result returned is 26/11/2005 - 02/03/2006......not sure about that second date, what result do you want in that case?
If you don't mind using analysis ToolPak functions then I suggest using this version utilising EOMONTH and EDATE, it should fix the above and also gets the correct results for your table of examples......but whether it works for all combinations of dates I'm not sure.....
=TEXT(C5+1,"dd/mm/yyyy")&" - "&TEXT(EDATE(B5,CEILING(DATEDIF(B5+1,EOMONTH(C5+1,0)+1,"m")+1,3)),"dd/mm/yyyy")
......or this version does exactly the same without add-in functions....
=TEXT(C5+1,"dd/mm/yyyy")&" - "&TEXT(MIN(DATE(YEAR(B5),MONTH(B5)+CEILING(DATEDIF(B5+1,DATE(YEAR(C5+1),MONTH(C5+1)+1,1),"m")+1,3)+{0,1},DAY(B5)*{1,0})),"dd/mm/yyyy")