I need to produce a query in cross-tab format that displays the column data as the first day of the week for every week for a given date range.

i.e. for August
4 Aug 03 | 11 Aug 03 | 18 Aug 03 | 25 Aug 03

Row1
Row2
etc...

Any pointer in the right direction would be appreciated.
Thanks
Tim

The basic idea is to make a query, which returns the value to be accumulated with the corresponding week indication, like:

SELECT YourGroupingField,
case datediff(wk, '2003-01-01', getdate()) < 30 THEN YourSumField ELSE 0 END AS Aug0,
case datediff(wk, '2003-01-01', getdate()) = 30 THEN YourSumField ELSE 0 END AS Aug1,
case datediff(wk, '2003-01-01', getdate()) = 31 THEN YourSumField ELSE 0 END AS Aug2,
case datediff(wk, '2003-01-01', getdate()) > 31 THEN YourSumField ELSE 0 END AS Aug3
FROM YourTable

You may use this query as a subquery or as a database view:

SELECT YourGroupingField, sum(Aug0), sum(Aug1), ....
FROM (YourQuery)
GROUP BY YourGroupingField

Let me know if this helps, or when you don't understand my outline.

Thank you for your quick reply.
I think I did not explain myself that well. So here goes - the column dates should auto-generate based on a date range. This date range could potentially be several months or years. So if the date range was 1 March 03 - 30 June 03 this would produde 18 columns starting with 3 Mar 03.

I know how to write a normal cross-tab but it's the auto-column generation that's tricky for me.

You want the heavy stuff? Look at this.