Forum Discussion
Why does Incremental Refresh require DateTime when data contains only date?
The best practice recommendation is valid for incremental refresh setups as well, guaranteeing that all time values inside a date will sit in the same partition. You can do the conversion at the last minute, via
let d = Datetime.From([Date value]) in d >= RangeStart and d < RangeEnd
Incremental Refresh expects a DateTime value that is usually pegged at midnight UTC, or a date integer. The DateTime requirement is a leftover from the OLAP cube partition days. While it is theoretically possible to specify partition boundary rules at a level lower than date (or on a totally different column) the default is that the smallest partition you can get is a daily partition (Always in UTC, mind you).
Just checking in on your earlier question regarding the use of Datetime columns for Incremental Refresh in Power BI.
lbendlin Irwan provided guidance explaining why Datetime is currently required (even for date-only data) and how appending a midnight timestamp is a valid workaround to balance cardinality and refresh behavior.
Could you please confirm if the provided solution helped in your scenario?
Let us know if you're still facing any blockers we're happy to assist further if needed!just circling back on your earlier question about using Datetime columns for Incremental Refresh in Power BI.
lbendlin Irwan had explained why a full Datetime column is required even when working with date-only data and suggested appending a midnight timestamp as a practical workaround to maintain refresh compatibility while keeping cardinality low.
Could you let us know if this approach worked for you?
If you're still running into any challenges, we're here to help!
