Fact design advice
2 posters
Page 1 of 1
Fact design advice
I am designing a transactional fact table that replicates the history of the customers transactions with us. Something along the following lines:
Based on this fact table design, if I have to create a report that shows me all the customers that moved from Value to Finest in month 03, would it be possible?
The report should show both, the ‘from’ product and the ‘to’ product along with the number every month.
Could I have a simpler fact table design? Maybe have two keys in the same row that shows ‘value’ and ‘finest' in the same row as ‘old prod key’ and ‘new prod key’?
But if I do that, then potentially every dimension could have an old value and a new value which will make the fact table very wide.
Customer | Product | Transaction Type | Month |
101 | Value | New Sale | 01 |
101 | Value | Moved From | 03 |
101 | Finest | Moved To | 03 |
102 | Value | New Sale | 01 |
102 | Value | Moved From | 03 |
102 | Finest | Moved To | 03 |
Based on this fact table design, if I have to create a report that shows me all the customers that moved from Value to Finest in month 03, would it be possible?
The report should show both, the ‘from’ product and the ‘to’ product along with the number every month.
Could I have a simpler fact table design? Maybe have two keys in the same row that shows ‘value’ and ‘finest' in the same row as ‘old prod key’ and ‘new prod key’?
But if I do that, then potentially every dimension could have an old value and a new value which will make the fact table very wide.
zipzap- Posts : 1
Join date : 2012-10-03
Fact design advice
Hi,
How many rows will be there in a month ? For one customer can you have only one transaction type or more ? WIill you have a report where you will compare data where the months are not consecutive.
The best is to join the data for two months and get the comparisn.
thanks
How many rows will be there in a month ? For one customer can you have only one transaction type or more ? WIill you have a report where you will compare data where the months are not consecutive.
The best is to join the data for two months and get the comparisn.
thanks
hkandpal- Posts : 113
Join date : 2010-08-16
Similar topics
» Advice on Fact Table Design
» FACT Design Question - How to design FACT Table when computation is required on the degenerate dimension?
» Design Advice
» I would like some advice regarding the following FACT table construction
» Modeling - Dimension/Fact - Need advice please
» FACT Design Question - How to design FACT Table when computation is required on the degenerate dimension?
» Design Advice
» I would like some advice regarding the following FACT table construction
» Modeling - Dimension/Fact - Need advice please
Page 1 of 1
Permissions in this forum:
You cannot reply to topics in this forum