Automated List Management

[Anonymous]

The goal is to automate my list managment process for software deployments.  

I am looking to create a calculated column, that will take a few filters and row count inter consideration.

For example.. I would first sort by user computer type > user location > user department.  Then I would say I want to break this up into waves of 50 people each.  So the output would be "Wave 1" for the first 50 users based on the sorting criteria, then as soon as it hit the 51st user, the next 50 users would be "Wave 2" 

Idealy I would want to be able to do this in Power BI so I can keep the direct connection to my data sources.  

The code for this column should also be able to update relatively easy.  For example, if I wanted to add sorts, or filter some things out, change the user count per wave from 50 to 100.. etc..

Any ideas would be super helpful!  

Thank you,

Ryan Mulhollem

[v-lili6-msft · Accepted solution]

HI, @Anonymous 

You could try this way:

First add an index column. Refer this post: https://community.powerbi.com/t5/Desktop/Creating-calculated-index-column-in-DAX/m-p/635397#M303922

Second, divide this index column by 50 (Integer)

INT((Table1[Index]-1)/50)

Then get result column

group = "Wave "&INT((Table1[Index]-1)/50)+1

For example:

In my simple sample, I use 3 instead of 50

Best Regards,

Lin

[Anonymous]

Hi @v-lili6-msft 

For the example, I changed the count change from 50 to 5 for simplistic reasons.  As you can see from below, the count is updating every 5, however if the location changes, the count resets.  I am getting thrown an error with the EARLIER function.. my formula is below.  Thank you again for your assistance here.

Test Column = COUNTROWS(FILTER(UserList, EARLIER(UserList[Location])= UserList[Location], INT((UserList[Index]-1)/5)+1))

UserName	Location	Expected Output
User 1	NYC	1
User 2	NYC	1
User 3	NYC	1
User 4	NYC	1
User 5	NYC	1
User 6	NYC	2
User 7	NYC	2
User 8	NYC	2
User 9	London	3
User 10	London	3
User 11	London	3
User 12	London	3
User 13	London	3
User 14	Baltimore	4
User 15	Baltimore	4
User 16	Baltimore	4
User 17	Baltimore	4
User 18	Baltimore	4
User 19	Baltimore	5
User 20	Baltimore	5
User 21	Baltimore	5
User 22	Baltimore	5
User 23	Houston	6
User 24	Houston	6
User 25	Houston	6
User 26	Princeton	7
User 27	Princeton	7
User 28	Princeton	7
User 29	Princeton	7
User 30	Princeton	7
User 31	Princeton	8
User 32	Princeton	8

---

---

---

[Anonymous]

@v-lili6-msft So technically this does serve as the answer to my question.  My only concern is the lack of customization.  

Is there a way to make this conditional as well?  For example - if I am doing wave 1 up to 50 people, however if the city changes (pre-sorted when making the index), the count should reset to 1.  So if there are only 38 people in NYC for example, the count should stop at the 38th person and reset at 39.  I know how to do this with Excel just not Power Query / BI

[v-lili6-msft]

hi, @Anonymous 

For your requirement, you need to use EARLIER Function to add the index column by group.

Best Regards,

Lin

[Anonymous]

Hi Lin,

First thank you for your response and help.  

The EARLIER Function only works on numeric values.  I want the conditional column to be based off of user location such as NYC, Baltimore, China, etc.. 

This is an example of my failed attempt..

[v-lili6-msft]

hi, @Anonymous 

EARLIER Function should be used as below:

Earlier(NonMHFTreeUsers[Location])=NonMHFTreeUsers[Location]

And I couldn't find out the logic of your formula, you could share some simple sample data and expected output.

Best Regards,
Lin

[Anonymous]

Thank you for your response Lin.  Let me work on this and let you know if it works!

[Anonymous]

Also, just a thought.  It would be useful if I could grab the category of another column for the naming convention of these migration waves.  For example, if the user is located in Asia Pacific, the wave name would be "Asia Pacific - Wave 1"