SQL Pivot on multiple aggregated columns

pivotmultiplecolumns.sql

DROP TABLE TempFacts

CREATE TABLE TempFacts (
	Model NVARCHAR(100)
	,DATE DATETIME
	,Engine NVARCHAR(100)
	,Revenue DECIMAL(18, 6)
	,Conversions DECIMAL(18, 6)
	)

--insert data
insert into Tempfacts
values('F','2/22/2017','bing',1,2),('F','2/22/2017','facebook',2,3),('F','2/22/2017','google',3,4),
('L','2/22/2017','bing',32,12),('L','2/22/2017','facebook',34,16),('L','2/22/2017','google',5,9),
('W','2/22/2017','bing',5.6,1.5),('W','2/22/2017','facebook',100.4,30.9),('W','2/22/2017','google',23.8,4.9)


SELECT EngineType
	,Sum(FConv) FConv
	,Sum(LConv) LConv
	,Sum(Wconv) Wconv
	,Sum(FRev) FRev
	,Sum(LRev) LRev
	,Sum(WRev) WRev
	,DATE
FROM (
	SELECT *
	FROM (
    -- Aggregate whatever we need
		SELECT DATE
			,sum(Conversions) Conversions
			,sum(Revenue) Revenue
    -- The below aliases, will become columns after pivot 
			,CASE 
				WHEN Model = 'F'
					THEN 'FConv'
				WHEN Model = 'L'
					THEN 'LConv'
				WHEN Model = 'W'
					THEN 'WConv'
				END AS ConversionModelType
			,CASE 
				WHEN Model = 'F'
					THEN 'FRev'
				WHEN Model = 'L'
					THEN 'LRev'
				WHEN Model = 'W'
					THEN 'WRev'
				END AS RevenueModelType
			,CASE 
				WHEN Engine = 'bing'
					THEN 'Bing'
				WHEN engine = 'google'
					THEN 'Google'
				WHEN engine = 'facebook'
					THEN 'Facebook'
				END AS EngineType
		FROM TempFacts
		GROUP BY Model
			,Engine
			,DATE
		) AS data
  -- Put aggregated values for Conversions under FConv,WConv,LConv 
	pivot(sum(conversions) FOR ConversionModelType IN (
				FConv
				,LConv
				,WConv
				)) AS pvt
  -- Put aggregated values for Revenue under FRev,WRev,LRev
	pivot(sum(revenue) FOR RevenueModelType IN (
				FRev
				,LRev
				,WRev
				)) AS pvt2
	) source
GROUP BY source.EngineType
	,DATE