Get certified for free when you join Fabric Data Days 2026 and dive into Fabric, Power BI, SQL, AI, and other essential data skills.
Join nowData Days is here! Join us now for 60+ days of learning, challenges, and connection. Learn more
I have a table A that looks like this:
| Time | Value | Service |
| 2022-01-01 | 10 | Service A |
| 2022-01-01 | 5 | Service B |
| 2022-01-02 | 8 | Service A |
| 2022-01-02 | 7 | Service B |
| 2022-01-03 | 4 | Service A |
| 2022-01-03 | 3 | Service B |
How do I calculate a column that takes the difference for Service A and Service B at each timestep? Either that the result is in a new column in Table A or in a entierly new Table B.
Solved! Go to Solution.
Hi @Anonymous ,
Easiest way is pivot table in Power Query like below:
let
Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("i45WMjIwMtI1MAQiJR0lQwMgEZxaVJaZnKrgqBSrgyZviiTthCptBJSywK0bJG2OW7cxUMoEt26QtDGq7lgA", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [Time = _t, Value = _t, Service = _t]),
#"Changed Type" = Table.TransformColumnTypes(Source,{{"Time", type date}, {"Value", Int64.Type}, {"Service", type text}}),
#"Pivoted Column" = Table.Pivot(#"Changed Type", List.Distinct(#"Changed Type"[Service]), "Service", "Value", List.Sum)
in
#"Pivoted Column"
Then you could directly create a column use [Service A] - [Service B].
Best Regards,
Jay
Hi @Anonymous ,
Easiest way is pivot table in Power Query like below:
let
Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("i45WMjIwMtI1MAQiJR0lQwMgEZxaVJaZnKrgqBSrgyZviiTthCptBJSywK0bJG2OW7cxUMoEt26QtDGq7lgA", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [Time = _t, Value = _t, Service = _t]),
#"Changed Type" = Table.TransformColumnTypes(Source,{{"Time", type date}, {"Value", Int64.Type}, {"Service", type text}}),
#"Pivoted Column" = Table.Pivot(#"Changed Type", List.Distinct(#"Changed Type"[Service]), "Service", "Value", List.Sum)
in
#"Pivoted Column"
Then you could directly create a column use [Service A] - [Service B].
Best Regards,
Jay
@Anonymous , refer this
Compare Categorical Data Using Slicers - Compare two Brands: https://youtu.be/exN4nTewgbc
Don't miss out on Data Days, June 15 through August 7. Learn Fabric, Power BI, SQL, AI and more.
Check out the May 2026 Power BI update to learn about new features.
| User | Count |
|---|---|
| 23 | |
| 21 | |
| 20 | |
| 18 | |
| 13 |
| User | Count |
|---|---|
| 58 | |
| 50 | |
| 38 | |
| 31 | |
| 27 |