The ultimate Fabric, Power BI, SQL, and AI community-led learning event. Save €200 with code FABCOMM.
Get registeredCompete to become Power BI Data Viz World Champion! First round ends August 18th. Get started.
Dear all
just to collect some thoughts if that is a good way or if this formula will slow down my pbi report in future...
I have two calculated values (Revenue/Units) and a calculated average (ASP). The user can choose from a button which value he wants to see. That works fine and fast with the switch function if i have one variable sum. Now i want to add targets as a second dimension, so i created the following formula:
var_Act vs. Target % = Switch( True();
values(dim_measureswitch[Measure]) = "Revenue"; [var_month_line_chart LC]/[var_Target LC]-1;
values(dim_measureswitch[Measure]) = "Units"; [var_month_line_chart LC]/[var_Target LC]-1;
values(dim_measureswitch[Measure]) = "ASP"; [var_month_line_chart LC]/[var_Target LC]-1;
blank())
var_Target LC = Switch( True();
values(dim_measureswitch[Measure]) = "Revenue"; [Target Sales LC];
values(dim_measureswitch[Measure]) = "Units"; [Target Unit];
values(dim_measureswitch[Measure]) = "ASP"; divide([Target Sales LC];[Target Unit];0);
blank())
var_month_line_chart LC = Switch( True();
values(dim_measureswitch[Measure]) = "Revenue"; ([Sales Revenue LC]-[MTD Sales Revenue LC])+ [MTD Sales Revenue LC RR CD];
values(dim_measureswitch[Measure]) = "Units"; ([Sales Units]-[MTD Sales Units])+ [MTD Sales Units RR];
values(dim_measureswitch[Measure]) = "ASP"; divide(([Sales Revenue LC]-[MTD Sales Revenue LC])+ [MTD Sales Revenue LC RR CD];([Sales Units]-[MTD Sales Units])+ [MTD Sales Units RR];0);
blank())
As you see i have now two variables (one with revenue (var_month_line_chart) and one with targets (var-target LC). That works fine, but the speed is decreasing.
What do you think is this the right way or is there a more efficient way to do this calculation?
KR
Florian
Hi @Flori_Abb,
From what I know, your formulas are nice, it's normal to run lower after you add more variables. I will post the update when I find more efficient way.
Best Regards,
Angelia
Hi @Flori_Abb
Not so much a performance tip, but I'd recommend using the DIVIDE function for your division maths.
so rather than
z = x / y
use
z = DIVIDE(x,y,0)
In terms of performance, can you preaggregate or precalculate some of your data upstream of Power BI? eg in the Query Editor or in your data source?
User | Count |
---|---|
82 | |
81 | |
35 | |
32 | |
32 |
User | Count |
---|---|
93 | |
79 | |
62 | |
54 | |
51 |