Forum Discussion

rougester's avatar
rougester
Regular Visitor
8 years ago

New to Power BI - Data & KPI Question

Hi,

 

The business I work for is going through system change and we have had our new data visualisation tools installed - Power BI!

 

I'm currently playing around with the package and have some questions regarding dataset.

 

...please bear with me, minimal experience and I've been thrown in at the deep end!

 

Regarding data, I have directed the desktop software at a basic sales database to get started and to practie with which includes columns for sales value, week number, year, store name.

 

My question, what is best practice for datasets? Would users tend to have multiple columns of 'sales value' for, say, each year?

 

For example, I'm trying to set up a KPI visual (I know, perhaps running before I can walk?) to show a +/- prior year variance. Do I need to create measures off the 'sales value' field which include year filters for the KPI visual to work?

 

This post might sound ridiculous but I appreciate any help offered.

 

Thanks,

M.

2 Replies

  • v-ljerr-msft's avatar
    v-ljerr-msft
    Microsoft Employee

    Hi rougester,


    For example, I'm trying to set up a KPI visual (I know, perhaps running before I can walk?) to show a +/- prior year variance. Do I need to create measures off the 'sales value' field which include year filters for the KPI visual to work?


    If I understand you correctly, you may need to create a measure to calculate the prior year sales value, so that you can use the measure as target on a KPI visual to show the variance on your report. :smileyhappy:

     

     

    Regards

    • rougester's avatar
      rougester
      Regular Visitor

      Thank you v-ljerr-msft

       

      I was a little unsure if my dataset was 'general' practice or not. It looks as though it is and that measures need to be created for variances.

       

      I'm getting issues with my KPI still, but I'll caryy on muddling through - practice is the key, right?!

       

      thanks again for you reply :)