Need page date Slicer to change 2 graphs, but each graph has different dates

[wokka]

I need to find out whether this is possible - we have a dimension table that has a column [First of Month] - which has data running from 01/01/2022,  01/02/2022,.....  up  to 01/05/2025. 

If we build a normal graph, we can choose the first of a month and this will fill in the data as needed. This is existing functionality and works well. All good.

But - I need to builld two powerbi graphs - one graph uses the chosen [First of month] value via slicer on the page  e.g.  01/10/2024 and a second graph needs to use a different [First of month] e.g. 01/09/2024. The two graphs need to sit side by side on the page and as I alter the [First of of month] slicer on the page, both graphs need to automatically change.

What I want to do is ( i think ) is create a synthetic column on the fly for 01/09/2024 for a graph. The issue is that we dont have this column in the existing calendar table for the [Preceding first of month] ( which would be ideal ) so I am hoping we can we do it, and if so, how could we do it please?

I have trialled setting up my second graph ( for the preceeding month -  01/09/2024 ) that is de-coupled from the page date slicer value , but we currently have to manually choose the first of month date, which we would love to avoid. 

Thank you in advance.

[v-venuppu]

Hi @wokka ,

Thank you @rohit1991 @Akash_Varuna for the prompt response.

May I ask if you have resolved this issue? If so, please mark the helpful reply and accept it as the solution. This will be helpful for other community members who have similar problems to solve it faster.

Thank you.

[rohit1991 · Accepted solution]

Hi @wokka ,

Yes, what you're aiming to achieve in Power BI is definitely possible by leveraging DAX and disconnected slicers. The core idea is to use a single slicer to drive both visuals, but have one graph display data for the selected month and the other show data for the preceding month—without requiring manual selection. Since your existing date table lacks a "Previous Month" column, you can simulate this by creating a measure that dynamically shifts the context.

First, use a disconnected date table (or a slicer based on a distinct list of [First of Month]) that does not filter your model directly. Then, for your two graphs, create two separate measures: one that uses the selected date as-is (e.g., with SELECTEDVALUE('DisconnectedDate'[First of Month])) and another that subtracts one month using EDATE (e.g., EDATE(SELECTEDVALUE('DisconnectedDate'[First of Month]), -1)). Within these measures, use CALCULATE with a FILTER that overrides the date context accordingly. This way, both visuals will update based on a single slicer selection—one showing the selected month, the other automatically showing the prior month—without needing to manually choose the second date.

Passionate about leveraging data analytics to drive strategic decision-making and foster business growth.

Connect with me on LinkedIn: Rohit Kumar.

[Akash_Varuna]

Hi @wokka  I think you could achieve this by creating a measure to calculate the preceding month dynamically using DATEADD, like PreviousMonthValue = CALCULATE(MAX(Calendar[First of Month]), DATEADD(Calendar[First of Month], -1, MONTH)). Use the slicer to directly filter the first graph, while the second graph references the preceding month using a measure such as FilteredDataForPreviousMonth.