How to display rows from datasets with variable schema in a single table visual?

Divyanshu1504 

You mean with a slicer selection? 

You can combine everything into a single table that includes all columns from all datasets, though some rows will have blanks where certain columns don’t apply. Add all columns to a field parameter and create another table to manage column visibility. This table will need a many-to-many, single-direction relationship with the field parameter table. The resulting view table will look like this.

After setting up everything, the output would be somehow look like the image below.

You might have to update this view table as new columns are added to/remove from the datasets. 

Another option is to unpivot all columns except the dataset column names, but this approach isn’t recommended because it can mix up data types, create complex relationships, and lead to complicated DAX.

Divyanshu1504  Hey,
I will refer below suggestion for this.

What’s possible (and What I will use in this case)

• Use an EAV/pivot-by-attribute pattern with a Matrix visual

- In Snowflake, FLATTEN the JSON into key/value rows
- Fields: dataset_name, record_id, attribute, value_text, value_num, value_dt
- Example: select dataset_name, record_id, f.key::string as attribute, to_varchar(f.value) as value_text, try_to_number(f.value) as value_num, try_to_timestamp(f.value) as value_dt from raw t, lateral flatten(input => t.payload) f;
- Create an Attribute Meta table: attribute, dataset_name, display_name, data_type, column_order, format
- Build a Matrix in Power BI:
- Rows: record_id
- Columns: Attribute Meta[display_name] (sort by column_order)
- Values: one measure that formats output based on Attribute Meta[data_type], e.g. SWITCH to pick value_num/value_dt/value_text and FORMAT accordingly
- Add a slicer for dataset_name. Columns will change automatically with the selected dataset’s attributes

Why this works

• Power BI visuals need a fixed field list, but a Matrix with “Columns = attribute” effectively creates dynamic columns driven by data (no need to predefine all columns)

Few more sugestion 

• Field Parameters can swap a predefined set of columns/tables, but won’t auto-adapt to new attributes/datasets (maintenance overhead)
• If you must export with fully dynamic columns, use Paginated Reports: hide/show columns by parameter and render only those present
• Keep types consistent via Attribute Meta. Example measure:

- Display Value =
VAR dt = SELECTEDVALUE('Attribute Meta'[data_type])
RETURN SWITCH(dt,
"NUMBER", FORMAT(MAX('Fact'[value_num]), SELECTEDVALUE('Attribute Meta'[format])),
"DATE", FORMAT(MAX('Fact'[value_dt]), "yyyy-MM-dd"),
MAX('Fact'[value_text])
)

• Performance: prefilter in Snowflake (dataset, date), ensure one row per record_id+attribute, and use DirectQuery only if necessary

Alternatives

• Show raw JSON using a JSON Viewer custom visual for drill-through details
• Paginated Report for email/export scenarios with truly dynamic columns

Bottom line for your problem

a) Best-practice in Power BI is: flatten JSON to EAV in Snowflake → Matrix visual with attribute on Columns → one formatting measure → dataset slicer for dynamic structure.

Thanks

Haish K 

If I resolve your issue. Kindly give kudos to this post and accept it as a solution so other can refer this.

Hi Divyanshu1504 ,

We haven’t received an update from you in some time. Could you please let us know if the issue has been resolved?
If you still require support, please let us know, we are happy to assist you.

Thank you.

Hi Divyanshu1504 ,

I wanted to check if you had the opportunity to review the information provided. Please feel free to contact us if you have any further questions.

Thank you.