Hi Dave,
Greetings back from Hyde, England :-)

Part of the problem is that, for example, they've sometimes used Column F
for Post Code (or ZipCode) & sometimes used it for the Post Town or County!
Here's some examples...
E F G
2 North Yorkshire HG4 1HJ
3 Rotherham South Yorkshire S63 7QQ
4 Beverley North Humberside HU17 8AZ
5 Sheffield South Yorkshire S10 2LN
6 South Yorkshire DN1 2DZ

So ideally, I want the "North" & "South" Yorkshires in one column,
Rotherham, Beverley & Sheffield in another & Postcodes in a third. But how to
do this programatically is, I believe, a bit of a nightmare!!

Regards
Colin
"Dave O" wrote:

Hi, Colin-
Greetings from Virginia, USA, near Washington DC.

This is always a tricky task, because finding a reliable pattern in
the data is difficult, as you're finding out. When you isolate the
post code out of the mix, you may find another pattern from there.

Do you mind posting some examples? If the post code always appears at
the end of the address, you might use and IF statement with a MID that
looks for a space 4 or 5 characters from the end, and isolate the post
code that way. If you post examples, for privacy reasons I suggest
substituting fake data.

Another technique that you may be able to use: you can use ASCII
character codes to look for upper and lower cases. For instance, you
can find the lower case "L" in "Colin" with this formula:
=SEARCH(CHAR(108),"Colin",1)
Character 108 is the ASCII representation of lower case L. One place
to find ASCII codes is
http://www.neurophys.wisc.edu/www/comp/docs/ascii.html

Dave O