Hi everyone,
We’re running into a recurring issue with user expectations around percentage totals in pivot tables, and I’m wondering if anyone has tips or workarounds to improve clarity.
We’re using a pivot table with multiple levels, subtotals, and totals. It includes:
- YoY value difference
- YoY % difference
QuickSight calculates totals by aggregating from the lowest-level data, which we understand – for % difference, it typically averages the underlying % values.
Users repeatedly report that the % total looks “wrong” because it doesn’t match what they get by doing a manual % = total diff / total base calculation.
They expect the % to be calculated directly from the displayed total values, not as an average of the rows.
We know QuickSight is technically doing the right thing (aggregating by column, averaging the % diff from each row), but from a user perspective, it’s confusing – especially when the total value sums correctly, but the % doesn’t “match” and when we have multiple levels as shown on second picture.
Appreciate any advice – even just how to better communicate this behaviour to users!
Thanks in advance.
