Question

I am creating a bucket on whether insurance was reported or not for my database.table per booking.

What i want to do is get a cound per userid where insurance reported is yes and a count per userid when insurance = no and then group by bmm, byyyy, dmm, dyyyy

it sounds simple but i think the bucket is confusing me

SELECT     CASE WHEN [insreported] IS NULL THEN 'NO' WHEN [insreported] = 'YES' THEN 'YES' END AS BUCKET, USERID, BMM, BYYYY, DMM, DYYYY
FROM         (SELECT     BOOKNO, USERID, INSREPORTED, BMM, BYYYY, DMM, DYYYY
                       FROM          dbo.agent_insurance_incentive_data) AS derivedtbl_1

Answer 1

You’re probably looking for something like this.

SELECT sum(yes),sum(no),BMM, BYYYY, DMM, DYYYY FROM (
    SELECT   SUM(  CASE WHEN [insreported] IS NULL THEN 0 WHEN [insreported] = 'YES' THEN 1 ELSE 0 END ) as yes,  SUM(  CASE WHEN [insreported] IS NULL THEN 1 WHEN [insreported] = 'YES' THEN 0 ELSE 1 END )as no, USERID, BMM, BYYYY, DMM, DYYYY
       FROM dbo.agent_insurance_incentive_data
       GROUP BY USERID, BMM, BYYYY, DMM, DYYYY 
    ) derived
GROUP BY BMM, BYYYY, DMM, DYYYY

The inner query counts the number of reported/not reported per user and BMM, BYYYY, DMM, DYYYY. The outer query sums this by user. I’m not quite sure what your goal is, but here this could be rewritten as just the outer query since the summation will be the same.