Forum Discussion
PowerBI - subtract two columns from different tables
Hi,
I need your help with the following problem. I have two different but similar tables from which I want to subtract two columns (Cost). For confidentiality reasons, I cannot provide the data but I will give a simple example:
Table A (imagine that I have multiple Product Clusters, Products, Years, etc. - this is a simplified view):
| Product Cluster | Product | Year | Attribute | Cost |
| A | X | 2023 | Sum | 10 |
| A | X | 2023 | Cont. | 0 |
| A | X | 2024 | Sum | 20 |
| A | X | 2024 | Cont. | 0 |
| A | X | 2025 | Sum | 30 |
| A | X | 2025 | Cont. | 1 |
| A | Y | 2023 | Sum | 10 |
| A | Y | 2023 | Cont. | 0 |
| A | Y | 2024 | Sum | 20 |
| A | Y | 2024 | Cont. | 0 |
| A | Y | 2025 | Sum | 30 |
| A | Y | 2025 | Cont. | 1 |
| A | Z | 2023 | Sum | 15 |
| ... | ||||
| B | H | 2023 | Sum | 20 |
| B | H | 2024 | Cont. | 0 |
Table B:
| Product Cluster | Product | Year | Attribute | Cost |
| A | X | 2023 | Sum | 9 |
| A | X | 2023 | Cont. | 0 |
| A | X | 2024 | Sum | 19 |
| A | X | 2024 | Cont. | 0 |
| A | X | 2025 | Sum | 35 |
| A | X | 2025 | Cont. | 1 |
| A | Y | 2023 | Sum | 12 |
| A | Y | 2023 | Cont. | 0 |
| A | Y | 2024 | Sum | 25 |
| A | Y | 2024 | Cont. | 0 |
| A | Y | 2025 | Sum | 20 |
| A | Y | 2025 | Cont. | 1 |
| A | Z | 2023 | Sum | 18 |
| ... | ||||
| B | H | 2023 | Sum | 10 |
| B | H | 2024 | Cont. | 0 |
Both come from the same Excel file and tab, with the only difference being the file version (version controlled on SharePoint). Many users can edit this file and I need to provide a quarterly review and capture the Delta.
Note that while transforming my data, I am filtering the data creating a new column to split the Sum from Cont. entries (Add Conditional Column filtering Sum).
What I have done is add an Index column to each table, create a 1-1 relationship and it worked. However, when I took another file version (same structure, same items on the same lines) it messed up the Indexing resulting in a mess.
I understand that the Indexing idea was not perfect as it cannot handle changes in line items - which can happen.
Is there anything else I can try?
Thanks,
George
2 Replies
- amitchandak
Super User
kg22 , you need to create common dimensions product, cluster product, attribute, year etc and join with both tables and then use them in visual
Power BI- Power Query: When I asked you to create common tables: https://youtu.be/PqfGW6pl1Sw
Power BI- DAX: When I asked you to create common tables: https://youtu.be/a2CrqCA9geM
https://medium.com/@amitchandak/power-bi-when-i-asked-you-to-create-common-tables-a-quick-dax-solution-8e3eccb41bda- kg22New Member
Thank you for the information. I've created the proposed structure and managed to get the required visuals i.e. the Delta of all costs between the two different tables.
This is done using the following Measure:
Delta (K€) = SUM('Table A'[Cost]) - SUM('Table B'[Cost]).
However, I have added a Table below this visual to present the entries contributing to this result i.e. those having a Delta not equal to "0". While filtering out all such entries, for some reason, I am seeing most of the entries reported with their negative value.
To give an example, if a 2023 item has 10 as Cost in Table A and it is the same in Table B, I would expect the result (Delta) to be 0. However, the outcome is -10.
This is a sample of the expected result:
Product Cluster Product Year Cost (Table A) Delta A X 2023 10 1 A X 2024 20 1 And what I see (the second shouldn't be appearing assuming the cost is the same in both tables):
Product Cluster Product Year Cost (Table A) Delta A X 2023 10 1 C L 2026 20 -20 However, the calculation should be ok as the visuals agree with the math (cross-checked in Excel).
Any idea why this is happening?
Thank you!
George