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.
Hi there,
I have a database that contains a list of users events in my App, basically where and when they click. The collumns I have are User ID, event type (as login, product selection, cart, payment options and confirmation) and time (in timestamp). I wanna know how many time users spend in each action using this data.
This can be done by creating a collumn that contains the time diference between two consectutive events of the same user (the first event can be ignored), but I coudn´t find a way to realize this. I have also tried to use the function EARLIER, but since my database is 20000 rows long, it keeps loading for minutes until I give up on it.
Can Power BI do this?
Hello @leonardosilva ,
Can you please share a sample data set along with the expected outcome?
Cheers!
Vivek
https://www.vivran.in/
Connect on LinkedIn
This is the desired outcome and an example I did with the EARLIER function:
When I applied this to the original database, it took 26 minutes to display a mensage saying I don´t have enough memory (I have 8GB) to make the process.
Hi @leonardosilva ,
Please try this one.
Column =
VAR t = 'Plan1'[timestamp]
VAR last =
CALCULATE (
MAX ( 'Plan1'[timestamp] ),
FILTER (
'Plan1',
'Plan1'[timestamp] < t
&& 'Plan1'[userid] = EARLIER ( 'Plan1'[userid] )
)
)
RETURN
t - last
Hi, @v-frfei-msft
I´ve tried this solution in my database but it also took 35 minutes and again displayed a mensage saying I don´t have enought memory. This is probably happening because my database is 20000+ rows long. Do you know another way to solve this using less memory?
User | Count |
---|---|
82 | |
81 | |
35 | |
32 | |
32 |
User | Count |
---|---|
93 | |
79 | |
62 | |
54 | |
51 |