I have a data set that has dates like so:
datetimecreated 2019-09-14 06:06:15.863383 2019-09-14 06:06:16.863385
When I go to edit my data set and force column datetimecreated into datatype date time, I get error:
Known date formats were not detected in this data. Provide a date format to transform this data into a known date format.
The data format I am trying to pass is this:
yyyy-MM-dd HH:mm:ss.ffffff
What am I doing wrong that this can not be detected as date field by Quicksight?
I've even tried this format:
yyyy-MM-dd HH:mm:ss
get same error as above.
First check would be if the date format is accepted in quicksight:
Quicksight User Guide pg.71
I don't think "yyyy-MM-dd HH:mm:ss:SSSSSS" is an accepted format.
In which case you may want to just extract the datetime data from the string to the most relevant format so that is usable for your analysis.
To do so you can create a calculated field based on your "datetimecreated" field using the parseDate() function:
parseDate(date, [format], [time_zone])
It's possible to create a calculated field either in the "Edit Data" section of the dataset by selecting it the dropdown menu for the "datetimecreated" field in the field list OR by selecting the option in the "Add" menu within a specific analysis.
Inserting your data into the formula would look something like this:
parseDate({datetimecreated}, yyyy-MM-dd HH:mm:ss, [time_zone])
This should create a date field with the extracted date information that you can then manipulate as with any other date and aggregate by DAY, MONTH etc.