Question

I’m pretty sure I need a pivot to do this but can’t quite figure it out. (sql newb)

I have data like this:

ID    CompanyID    Year    Revenue    Expenses
1     0003         2011    12000      4000
2     0003         2010    9000       6000
3     0003         2009    7000       9000
4     0010         2011    134300     34000
5     0010         2010    43000      46000
6     0010         2009    73000      39000

Can I use a Pivot to display this table like this:

CompanyID    2011-Revenue    2010-Revenue    2009-Revenue    2011-Expenses   2010-Expenses    2009-Expenses
0003         12000           9000            7000            4000            6000             9000
0010         134300          43000           73000           34000           46000            39000

Here is what I have so far…

SELECT P1.*
FROM    (SELECT [CompanyID]
            ,CASE P.[Year] WHEN 2011 THEN P.[Revenue] ELSE NULL END AS '2011-Revenue'
            ,CASE P.[Year] WHEN 2010 THEN P.[Revenue] ELSE NULL END AS '2010-Revenue'
    FROM tblRecords P WHERE P.[CompanyID] = @companyID GROUP BY CompanyID, [Year], [Revenue]) AS P1

Which is returning:

CompanyID    2011-Revenue    2010-Revenue 
0003         12000           NULL
0003         NULL            9000

Few problems with my results…

1. The there is two records for CompanyID 0003 I’d like it to group into one record

2. I can only choose 1 company at a time, I need to choose multiple. I tried

   FROM tblRecords P
   WHERE P.[CompanyID] IN (@CompanyIDs)
   GROUP BY CompanyID, [Year], [Revenue]) AS P1
   

   Where @CompanyIDs is a string like '0003, 0010' – It didn’t fail but the result was just an empty table with the headers and no data..

Any help would be appreciated.. or let me know if I’m misunderstanding pivot?

Thanks a lot!

Thomas

EDIT: Using Microsoft SQL Server Management Studio 2005 Express

UPDATE 2: I’ve figured out joining tables for more details however I still need to be able to pass in the the CompanyIDs as a comma delimited string.. any help on that would be appreciated.

vvvvvvvv I’VE FIGURED OUT BELOW THIS (will post once all is working) vvvvvvv

UPDATE: It looks like what Ruben has proposed is going to work however I’ve just determined I need a bit more functionality to this… Can I join this with another table to have the headers

  CompanyID    CompanyName    CompanyAddress    2011-Revenue    2010-Revenue 

Where CompanyName and CompanyAddress come from another table (tblCompanyDetails)

I’ve tried using:

SELECT *
FROM 
(
    SELECT  CompanyID, tblCompanyDetails.CompanyName, tblCompanyDetails.CompanyAddress, CAST(YEAR AS varchar) + ' - Revenue' Type,
            Revenue Value FROM tblRecords
    FROM tblRecords INNER JOIN tblCompanyDetails ON tblRecords.CompanyID = tblCompanyDetails.CompanyID
    UNION ALL
    SELECT  CompanyID, tblCompanyDetails.CompanyName, tblCompanyDetails.CompanyAddress, CAST(YEAR AS varchar) + ' - Expenses' Type,
            Expenses  Value FROM tblRecords 
    FROM tblRecords INNER JOIN tblCompanyDetails ON tblRecords.CompanyID = tblCompanyDetails.CompanyID
) src
    PIVOT
(
    SUM(Value) for [Type] in
([2011 - Revenue], [2010 - Revenue], [2009 - Revenue],
     [2011 - Expenses], [2010 - Expenses], [2009 - Expenses]
)
) pvt
WHERE CompanyID = @CompanyID

I get the error:

Msg 1038, Level 15, State 4, Procedure spCompare, Line 10
An object or column name is missing or empty. For SELECT INTO statements, verify each column has a name. For other statements, look for empty alias names. Aliases defined as "" or [] are not allowed. Add a name or single space as the alias name. 

Answer 1

Try this:

DECLARE @CompanyIDs XML
DECLARE @Records  TABLE
(
    ID int,
    CompanyID char(4),
    Year int,
    Revenue decimal,
    Expenses decimal
)
DECLARE @CompanyDetails TABLE
(
    CompanyID char(4),
    Name varchar(50),
    Address varchar(50)
)

SET @CompanyIDs = '
<filter>
    <CompanyID>0003</CompanyID>
    <CompanyID>0010</CompanyID>
</filter>'

INSERT INTO @Records 
(ID, CompanyID, Year, Revenue, Expenses)VALUES 
(1, '0003', 2011, 12000 , 4000 ),
(2, '0003', 2010, 9000  , 6000 ),
(3, '0003', 2009, 7000  , 9000 ),
(4, '0010', 2011, 134300, 34000),
(5, '0010', 2010, 43000 , 46000),
(6, '0010', 2009, 73000 , 39000)

INSERT INTO @CompanyDetails 
(CompanyID, Name, Address) VALUES
('0003', 'Company A', 'A Street'),
('0010', 'Company B', 'B Street')

SELECT *
FROM 
(
    SELECT  CompanyID, CAST(YEAR AS varchar) + ' - Revenue' Type,
            Revenue Value FROM @Records 
    UNION ALL
    SELECT  CompanyID, CAST(YEAR AS varchar) + ' - Expenses' Type,
            Expenses  Value FROM @Records 
) src
PIVOT
(
    SUM(Value) for [Type] in
    ([2011 - Revenue], [2010 - Revenue], [2009 - Revenue],
     [2011 - Expenses], [2010 - Expenses], [2009 - Expenses]
    )
) pvt
JOIN    @CompanyDetails Details
    ON  Details.CompanyId = pvt.CompanyID
WHERE   pvt.CompanyID IN
(
    SELECT  T.C.value('.', 'char(4)')
    FROM    @CompanyIDs.nodes('/filter/CompanyID') T(C)
)

You need to send your company filter as XML and use that subselect to break your data as required for pivot operations. For JOIN operations, just use pvt output