Forum Discussion

marzio's avatar
marzio
Frequent Visitor
7 years ago

Visualising best time to call

Hello,   I have a large dataset of calls of a call center grouped via a unique identifier ID. Each row has a single ID and (among many others) multiple columns with the time stamp of each call att...
Need Help
v-xicai's avatar
v-xicai
Icon for Community Support rankCommunity Support
7 years ago

Hi marzio ,

 

You may need to unpivot the same call type , call timestamp and call outcome into a separate columns.

 

For example, enter into Query Editor, click on columns call 1 timestamp - call N timestamp at the same time, then choose "Unpivot Columns" option, you can rename this result column "Attribute" with "call timestamp" . Don't forget to click the "Close & Apply" button. Then you can create relationship between the source table and the data to time intervals table on date field.

 

 

 

 

Best Regards,

Amy

 

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

marzio's avatar
marzio
Frequent Visitor
7 years ago

Hi Amy,

 

The Unpivot solution is going to the right direction but doing it for the data source multiplies every row, and that causes many issues to the rest of the data. All the charts and calculations are showing wrong data so I had to backtrack.

 

I have tried to SUMMARIZE the columns I need into a separate table but I cannot UNPIVOT that because it is a DAX function.

 

The step I am missing to make it work is creating a new table in Power Query based on a selection of columns.

 

So I can UNPIVOT the columns and follow the rest of your suggestion.

 

Any ideas for that step?

 

 

I got lost in other threads without finding a working solution.

 

Marzio