Forum Discussion
Paginated Report issue
- 23 days ago
Hi Deeva,
Since the tablix checks did not fix it, I would look at what is different between the two filter contexts rather than the tablix itself. Two things usually explain "only Sales, Sales Volume % and YOY are off":
1. Filters that exist in the Power BI report but not in the paginated report
The Power BI visual is often affected by report, page or visual level filters (or a slicer on a fact column) that are not part of the measure. The paginated report only sees what is in its own DAX query and parameters. Check the Filters pane in the PBI report, and use Performance analyzer > Copy query on the visual to see the exact DAX it runs.
2. The grain of the paginated dataset
The query designer builds one SUMMARIZECOLUMNS at the row level you selected. Ratio and YOY measures are only correct at the grain they were evaluated at, so semester or total cells built from them in the tablix will not match Power BI. A quick way to compare is to run both queries in DAX query view in Desktop:
// 1) paste the query copied from Performance analyzer (Power BI visual) // 2) paste the paginated dataset query (Dataset properties > Query) // then compare one account side by side: EVALUATE CALCULATETABLE ( SUMMARIZECOLUMNS ( 'Date'[Year], 'Date'[Semester], 'Date'[Month], "Sales", [Sales], "Sales Volume %", [Sales Volume %], "YOY %", [YOY %] ), 'Account'[Account ID] = "<one account>" )If the Power BI query has extra filters (KEEPFILTERS / TREATAS blocks) that the paginated one lacks, that is your difference. If semester values differ but months match, return the semester-level values from the model instead of aggregating months in the tablix, for example a second dataset grouped by Semester, or a dataset that already includes subtotals:
EVALUATE SUMMARIZECOLUMNS ( ROLLUPADDISSUBTOTAL ( 'Date'[Semester], "IsSemesterTotal", 'Date'[Month], "IsMonthTotal" ), 'Account'[Account ID], "Sales", [Sales], "Sales Volume %", [Sales Volume %], "YOY %", [YOY %] )So my overall approach would be:
Copy the Power BI visual query (Performance analyzer) -> Run it and the paginated query side by side for one account -> Different filters? add them as paginated parameters/filters -> Same filters but totals differ? return subtotal rows from DAX instead of Sum() in the tablixIf you can share the measure definitions for Sales Volume % and YOY, it would be easier to tell which of the two it is.
AI-assisted drafting: AI was used to help structure and phrase this response. I reviewed and validated the technical content before posting.
Hi Deeva,
Since the tablix checks did not fix it, I would look at what is different between the two filter contexts rather than the tablix itself. Two things usually explain "only Sales, Sales Volume % and YOY are off":
1. Filters that exist in the Power BI report but not in the paginated report
The Power BI visual is often affected by report, page or visual level filters (or a slicer on a fact column) that are not part of the measure. The paginated report only sees what is in its own DAX query and parameters. Check the Filters pane in the PBI report, and use Performance analyzer > Copy query on the visual to see the exact DAX it runs.
2. The grain of the paginated dataset
The query designer builds one SUMMARIZECOLUMNS at the row level you selected. Ratio and YOY measures are only correct at the grain they were evaluated at, so semester or total cells built from them in the tablix will not match Power BI. A quick way to compare is to run both queries in DAX query view in Desktop:
// 1) paste the query copied from Performance analyzer (Power BI visual)
// 2) paste the paginated dataset query (Dataset properties > Query)
// then compare one account side by side:
EVALUATE
CALCULATETABLE (
SUMMARIZECOLUMNS (
'Date'[Year], 'Date'[Semester], 'Date'[Month],
"Sales", [Sales],
"Sales Volume %", [Sales Volume %],
"YOY %", [YOY %]
),
'Account'[Account ID] = "<one account>"
)If the Power BI query has extra filters (KEEPFILTERS / TREATAS blocks) that the paginated one lacks, that is your difference. If semester values differ but months match, return the semester-level values from the model instead of aggregating months in the tablix, for example a second dataset grouped by Semester, or a dataset that already includes subtotals:
EVALUATE
SUMMARIZECOLUMNS (
ROLLUPADDISSUBTOTAL ( 'Date'[Semester], "IsSemesterTotal", 'Date'[Month], "IsMonthTotal" ),
'Account'[Account ID],
"Sales", [Sales],
"Sales Volume %", [Sales Volume %],
"YOY %", [YOY %]
)So my overall approach would be:
Copy the Power BI visual query (Performance analyzer)
-> Run it and the paginated query side by side for one account
-> Different filters? add them as paginated parameters/filters
-> Same filters but totals differ? return subtotal rows from DAX
instead of Sum() in the tablixIf you can share the measure definitions for Sales Volume % and YOY, it would be easier to tell which of the two it is.
AI-assisted drafting: AI was used to help structure and phrase this response. I reviewed and validated the technical content before posting.