Forum Discussion
Data row count changes after export
Hi I have imported the data into power bi desktop. After import when i put the same data into table it and extracted the same. The row count reduces. When i count the rows using "countrow" the count is correct but when i export the file it gets reduced.
The total row count is : 4198 Correct One !
After extract from table : 3970 Reduced One !
I have cross checked following things
- I have not applie any filter.
- Created only table having one column in it.
- Data type is Whole Number
- No erros.
- Perfored this with Matrix but it removes duplicates i want to keep them.
Then you can add an index column to your data and add it to the exported table. That should prevent the duplicates being removed.
Anyway using Power BI as an "Excel" exporter is not really what it's built to do.
If you really need to use the data in Excel, use Analyze in Excel instead.
6 Replies
mesuyog , if you are checking the number of rows vs row count measure, then you should sum up and check the row count measure after download, Most of visuals in power bi group the data by default. Means duplicate values(set of values) are not repeated
Thanks for the solution....but ...No i don't want to do any operations on the data just want to export as excel.....i have used COUNTROW just to crosscheck ...
If you put a number field into matrix "Values" section, it will automatically perform the defined aggregation (typically SUM). If you put it in the rows or columns section, it will remove duplicates. That's a standard and expected behavior.
You can use the standard table and set the field to "Do not summarize".
Thanks for the solution....but I have used Table only and made it do not summarize ... still its giving the same issue...
Then you can add an index column to your data and add it to the exported table. That should prevent the duplicates being removed.
Anyway using Power BI as an "Excel" exporter is not really what it's built to do.
If you really need to use the data in Excel, use Analyze in Excel instead.
