SQL Server CASE Statement in GROUP BY

People are often scared to put case statements in their GROUP BY clause because it could impact query performance. In this article by Grant Fritchey, he shows a few examples of how this might or might not work well, as well as how to verify if you have it working correctly.

You can click on that to expand it into something readable. We can eliminate the Parameter Sniffing from the equation if we want to by modifying the query thus:

CREATE PROCEDURE dbo.InvoiceGrouping_NoSniff (@x INT)
AS
DECLARE @x2 INT;
SET @x2 = @x;
 
SELECT  SUM(il.UnitPrice),
        COUNT(i.ContactPersonID),
        COUNT(i.AccountsPersonID),
        COUNT(i.SalespersonPersonID)
FROM    Sales.Invoices AS i
JOIN    Sales.InvoiceLines AS il
        ON il.InvoiceID = i.InvoiceID
GROUP BY CASE WHEN @x2 = 7 THEN i.ContactPersonID
              WHEN @x2 = 15 THEN i.AccountsPersonID
              ELSE i.SalespersonPersonID
         END;
GO

However, except for some deviation on the estimated rows (since it’s averaging the rows returned), the execution plan is the same.