DateHourMinute Dimension table

Hello everyone,

How have you guys solved the problem of a fact table having a grain of every 15 minutes?

I’m working on a Air quality dataset project - the grain of the client’s data by 15minutes, example 15/04/2020 08:15:00 // 15/04/2020 08:30:00 etc. We currently have a date dimension with just the DATE being the grain of the table. I’m thinking of creating another Dimension “HourMinute” with the grain being every 15minutes. But also considering creating a new dimension “DateHourMinute” with the grain being every 15 minutes of a given date. Example: 5/04/2020 00:00:00 // 5/04/2020 00:15:00 // 5/04/2020 00:30:00 // 5/04/2020 00:45:00 // etc …

How have you guys solved this type of problem please and what are your views?

Thanks heaps in advance


I’ve never dealt with this specific problem myself, but just saw this article earlier today by Reza Rad, which deals with this question directly:

I hope this is helpful.


Awesome, thanks heaps @BrianJ I’ll check it out.