Try using a database trigger

CREATE trigger mytrigger on database for create_table,alter_table,drop_table

AS

...

Both INSERT and UPDATE support the OUTPUT clause

UPDATE Target

SET Target.AlphaValue = Source.AlphaValue, Target.NumericValue = Source.NumericValue

OUTPUT inserted.*, deleted.*

FROM TargetDB.dbo.AdmFormData AS Target

INNER JOIN SourceDB.dbo.AdmFormData AS Source ON (Target.FormType = Source.FormType and Target.KeyField...

Use a separate INSERT and UPDATE statement, something like this

UPDATE Target

SET Target.AlphaValue = Source.AlphaValue, Target.NumericValue = Source.NumericValue

FROM TargetDB.dbo.AdmFormData AS Target

INNER JOIN SourceDB.dbo.AdmFormData AS Source ON (Target.FormType = Source.FormType and Target.KeyField...

I get a different error message

The insert column list used in the MERGE statement cannot contain multi-part identifiers. Use single part identifiers instead.

Try this instead

MERGE dbo.AdmFormData AS Target

USING dbo.AdmFormData AS...

Something like this??

DECLARE @t TABLE(DeliveryDate DATE,Assigned INT,Unassigned INT, Total INT)

INSERT INTO @t(DeliveryDate,Assigned,Unassigned)

VALUES

('2013-04-29',1,0),

('2013-04-29',1,1),

('2013-04-29',1,1),

('2013-04-29',1,1),

('2013-04-30',0,0),

('2013-04-30',0,1),

('2013-05-01',0,0);

SELECT DeliveryDate,Assigned,Unassigned,

SUM(Assigned + Unassigned) OVER(PARTITION BY DeliveryDate) Total

FROM @t

ORDER BY DeliveryDate

SELECT

(SELECT 'Field1' AS "@name", Field1 AS "text()" FOR XML PATH('ColumnValue'),TYPE),

(SELECT 'Field2' AS "@name", Field2 AS "text()" FOR XML PATH('ColumnValue'),TYPE),

(SELECT 'Field3'...

If you're using SQL Server 2012, you can do this

;with Data(ClassID, MemberID, Weight) as

(

select 'A', 1, 10 union all

select 'A', 2, 20 union all

select 'A', 3, 40 union all

select 'A',...

Also this

SELECT child.value('(parent::node()/ID/text())[1]', 'NVARCHAR(50)') AS [ID]

, child.value('(State/text())[1]', 'NVARCHAR(50)') AS [State]

, child.value('(Value/text())[1]', 'NVARCHAR(50)') AS [Value]

FROM [XmlTable]...

That...

This uses a tally table, probably not very efficient though.

DECLARE @Plates TABLE(StartPlate INT, StopPlate INT);

INSERT INTO @Plates(StartPlate, StopPlate)

VALUES(1,12),(15,23);

DECLARE @BrokenPlates TABLE(StartPlate INT, StopPlate INT);

INSERT INTO @BrokenPlates(StartPlate, StopPlate)

VALUES(3,7),(15,18),(21,22);

WITH E1(N) AS (

...

Maybe this?

SELECT d.ID

FROM DATA d

LEFT OUTER JOIN NEW_DATA n ON n.CODE = d.CODE AND n.CNAME=d.CNAME

GROUP BY ID

HAVING COUNT(*) = COUNT(n.CNAME);

Try changing

INSERT INTO Build_list

(ContractID, SectionID, ItemID, Qty)

SELECT...

Another one to try

WITH Numbers(N) AS (

SELECT N

FROM ( VALUES(1),(2),(3),(4) ) Numbers(N)),

Recur(N,Combination) AS (

SELECT N, CAST(N AS VARCHAR(1000))

FROM Numbers

UNION ALL

SELECT n.N,CAST(r.Combination + ',' + CAST(n.N AS VARCHAR(10)) AS VARCHAR(1000))...

You can do this instead, although I don't think it will make much difference

WITH Results AS (

SELECT CustomerName, PostalCode

FROM @CustomerOffices

WHERE OfficeId BETWEEN @FromOfficeId AND @ToOfficeId)

SELECT 'Summary' as Summary,

...

Good job of posting sample data + results. Thanks.

SELECT 'Summary' as Summary,

STUFF((SELECT ',' + CustomerName AS "text()"

FROM @CustomerOffices

...