question

This will give you the answer but it's not particularly elegant and I'm sure there is a more efficient way of doing it. SELECT * FROM Table WHERE Product NOT IN (SELECT Product FROM Table WHERE date_1 = '01-01-2017') AND Product NOT IN (SELECT Product FROM Table WHERE date_1 = '02-01-2017')

This will give you the answer, but it's not necessarily the right answer: DECLARE @ProductTable TABLE ( Date DATE, Product CHAR(2) ); INSERT INTO @ProductTable VALUES ('20170101', 'P1'), ('20170101', 'P2'), ('20170102', 'P2'), ('20170102', 'P3'); SELECT Product FROM @ProductTable GROUP BY Product HAVING COUNT(*) < 2; If you want it to do those which haven't sold on all dates in the set, given you were saying about multiple products, then perhaps changing the HAVING clause to: HAVING COUNT(*) < (SELECT COUNT(DISTINCT Date) FROM @ProductTable) At least this won't need maintenance when you change the date range.

I like Thomas's answer, and as he says it covers sales for "***all dates in the set***", but there's an edge case that I don't know if you care about or need. If the data has gaps, e.g. DECLARE @ProductTable TABLE ( Date DATE, Product CHAR(2) ); INSERT INTO @ProductTable VALUES ('20170101', 'P1'), ('20170101', 'P2'), ('20170103', 'P2'), ('20170103', 'P3'); and so all products should be returned as there are no sales of anything on 2 Jan, then you may need something like: SELECT Product FROM @ProductTable GROUP BY Product HAVING COUNT(*) < (select datediff(day, min(date), max(date))+1 from @ProductTable );