TLDR: You can use ISNULL when dealing with Pivoted data.
I was working with data reflecting feedback our customers have given on how well we did in 3 categories. A customer can provide any combination of those categories (i.e. just 1, 2 out of 3, or all 3). In order to make the data easier to work with for my internal customer, I decided to pivot the feedback and expose each category as a column.
However, we needed to account for NULL values, as the consuming tool apparently doesn't handle NULL data very well. Luckily, I discovered the following is completely possible and valid:
I was working with data reflecting feedback our customers have given on how well we did in 3 categories. A customer can provide any combination of those categories (i.e. just 1, 2 out of 3, or all 3). In order to make the data easier to work with for my internal customer, I decided to pivot the feedback and expose each category as a column.
However, we needed to account for NULL values, as the consuming tool apparently doesn't handle NULL data very well. Luckily, I discovered the following is completely possible and valid:
Code:
SELECT
foo
, ISNULL([lorem], '') AS lorem
, ISNULL([ipsum], '') AS ipsum
FROM
table1
PIVOT
(
AVG(foo) FOR bar IN ([lorem],[ipsum])
) AS pivottable
Last edited: