Forum Discussion
Why filter does not propogate
Hi mp390988 ,
Thank you for reaching out to Microsoft Fabric Community Forum.
Deku rosha_rosha Thank you for your quick response.
When you expand a customer in the matrix visual, Power BI filters the data based on that customer using the relationship between DimCustomer and FactInternetSales. This filter flows into the fact table and restricts the rows to only those related to that customer. From there, Power BI identifies the ProductKeys associated with that customer's transactions. Using those keys, it retrieves the corresponding product names from DimProduct. Even though there’s no direct relationship from DimCustomer to DimProduct, Power BI doesn’t need one — it leverages the fact table to bridge the two. That’s why you only see the products actually purchased by each customer, rather than a full list of all products.
If this post helps, then please consider Accepting as solution to help the other members find it more quickly, don't forget to give a "Kudos" – I’d truly appreciate it!
Regards,
B Manikanteswara Reddy
Filters only progate from one to many. So will no travel upstream to your dim product table. That is unless you have a bidirectional relationships, which is not best practice. You can update the measure as follows
Count of EnglishProductName =
CALCULATE(
COUNTROWS(
VALUES(DimProduct[EnglishProductName])
),
CROSSFILTER(
DimProduct[ProductKey],
FactInternetSales[ProductKey],
Both)
)
This will enable bidirectional relationships for that measure allowing the filter to propogate upstream
Hi,
thank you for your explanation. I understand filters propagate from the 1 side to the many side.
But now I don't understand why inside the first visual products are being filtered respective of customer name?
I would expect a complete list of product name across all customers as there is no relationship between Customer and Products and there is not a 1 to many relationship between Customer and Products.
i remember posting a question a while back regarding customer count per year. The model was the same one as in this question.
then I tried to show count of customers per year but it was repeating the same values:
So I would have expected the same behaviour in my very first visual where I have a matrix showing customer name and underneath each customer name I have product name. I would have expected repeating values of all product names just like how I'm seeing repeating count of customers for each year.
The underlying summarizecolumn is different for the two. In the matrix is it performs a autoexist from the fact table. This doesn't occur with the measure. If you had measure in the matrix for count of year you would see 6 for every combo of product and customer
Thank you for your reply but you completely lost me now.
Hi mp390988 ,
Thank you for reaching out to Microsoft Fabric Community Forum.
Deku rosha_rosha Thank you for your quick response.
When you expand a customer in the matrix visual, Power BI filters the data based on that customer using the relationship between DimCustomer and FactInternetSales. This filter flows into the fact table and restricts the rows to only those related to that customer. From there, Power BI identifies the ProductKeys associated with that customer's transactions. Using those keys, it retrieves the corresponding product names from DimProduct. Even though there’s no direct relationship from DimCustomer to DimProduct, Power BI doesn’t need one — it leverages the fact table to bridge the two. That’s why you only see the products actually purchased by each customer, rather than a full list of all products.
If this post helps, then please consider Accepting as solution to help the other members find it more quickly, don't forget to give a "Kudos" – I’d truly appreciate it!
Regards,
B Manikanteswara Reddy