...for that value to be stored in a variable...

Use sp_executesql like this

DECLARE @result varchar(255)

SET @sql =...

Then add a GROUP BY and HAVING like

SELECT loi.LOB_ID, loi.EUS_ID, loi.LOI_START_DATE, loi.LOI_END_DATE, eu.CTY_CODE

FROM @learning_object_instances loi

JOIN @end_users eu ON eu.EUS_ID = loi.EUS_ID

WHERE eu.CTY_CODE = 'RU' and loi.LOI_START_DATE...

DATEADD(day, -1, '31/10/2006')

Agree, requirements are unclear

As far as I can determine

SELECT i.LOB_ID, i.EUS_ID, i.LOI_START_DATE, i.LOI_END_DATE, u.CTY_CODE

FROM learning_object_instances i

CROSS JOIN end_users u

WHERE u.CTY_CODE = 'RU' AND i.LOI_START_DATE > 'September 7, 2006'

GROUP...

Not withstanding Jeff's comments

if all the names are the same format

PARSENAME(REPLACE(REPLACE(full_name,',',''),' ','.'),2) AS [First_Name],

PARSENAME(REPLACE(REPLACE(full_name,',',''),' ','.'),1) AS [Middle_Name],

PARSENAME(REPLACE(REPLACE(full_name,',',''),' ','.'),3) AS [Last_Name]

10-PATINDEX('%[^0]%',REVERSE(RIGHT(CAST(mynumber as varchar),9))+'1')

1,000,000 rows <=10 secs

This type of problem has occurred in a number of threads on this site with various answers, one of which is to do calculus on the days of the week by...

DATEDIFF(hour, date1, date2) - (DATEDIFF(week, date1, date2) * 48)

providing date1 and date2 are not sat/sun dates

Don't know what coding that is but my guess for a solution would be

CREATE PROCEDURE TopCustomerValue2

  @TopCompany CHAR(36) OUTPUT,

  TopValue INT OUTPUT)

AS

SELECT TOP 1 @TopCompany =...

In that case go back to my original solution and expand it

DECLARE @index int

SET @String = REPLACE(@String,' ','')

SET @index = PATINDEX('%Period:[0-9][0-9/][0-9/]%',@String)

SET @String = STUFF(@String,1,@index + 6,'')

SET...

OK then

SELECT

CAST(LEFT(STUFF([String],1,PATINDEX('%Period: [0-9/ ][0-9/ ][0-9/ ]%',[String])+7,''),PATINDEX('%[^0-9/ ]%',STUFF([String],1,PATINDEX('%Period: [0-9/ ][0-9/ ][0-9/ ]%',[String])+7,''))-1) as datetime) AS [FirstDate],

CAST(LEFT(SUBSTRING([String],PATINDEX('%Period: [0-9/ ][0-9/ ][0-9/ ]%',[String])+7+PATINDEX('%[^0-9/ ]%',STUFF([String],1,PATINDEX('%Period: [0-9/ ][0-9/ ][0-9/ ]%',[String])+7,''))+2,255),PATINDEX('%[^0-9/ ]%',SUBSTRING([String],PATINDEX('%Period: [0-9/ ][0-9/ ][0-9/...

I agree with Vladan that it  is over complicated

I believe this would achieve the same

SET DATEFORMAT DMY

SET @String = REPLACE(@String,' ','')

SET @String = SUBSTRING(@String,PATINDEX('%to[0-9][0-9/][0-9/]%',@String)+2,255)

SET @String = LEFT(@String,PATINDEX('%[^0-9/]%',@String)-1)...

...don't have the budget to create a data warehouse...

Know that as well, but if you could ever get...

Some table defs, sample data and actual required results would be helpful

But I consulted my Oiuja board  and.....

SELECT [Category],[ProductName],[CompanyName]

FROM

(SELECT 1 AS [Category],p.[ProductName],p.[CompanyID]

FROM [Product] p...