vendredi 5 juin 2015

Designing tables for measures efficiently

I have a fairly good understanding of how to normalize my data for employees, management, locations, etc, but I can't figure out a good way to normalize performance metrics?

Datetime | Name | ManagerID | LocationID | ... | Surveys | Rating | Interactions | Duration

Should each measure be a separate column on the same table?

How do you keep track of targets which may change frequently (monthly)? Is the target for each metric also a column (this can't be or it would repeat constantly for each data and employee).

Some measures are 'required' and others are just additional information. is there a way to separate the required measures into their own table? I would need to report on the required metrics ultimately, but have access to the 'non-required' metrics to drill down.

I can't seem to find any best practices. I am dealing with measures which vary per project and not all measures are required, and some have frequent target changes so it is difficult to assess goal attainment (actual/target). Can someone point me to the right direction?

Aucun commentaire:

Enregistrer un commentaire