Hey All,
So... I have an evar dimension with lots of customer classifications in place. It's a monthly automation/feed setup in data sources that sends about 12 columns/classifications of user/visitor attributes (think like customer account data). I found out that this automation is missing some important classifications for our customers, so I want to do a manual import of about 5-10 more classifications. Does anyone know if this can break the automated data sources data that is coming in? Curious because whenever you create new sub-classifications or sub-dimensions, it updates the template for that evar (with about 5-10 more classifiations). I wasn't sure if it will mess up the template for that evar for data sources and hope this makes sense! When I do the classificaiton import manually, I'll leave all the other columns blank (existing/older classifications) so it doesn't touch any of those and overwrite by accident. That said, I just wasn't sure if you add a bunch of new classifications for a single evar that already has data sources setup- if new classifications will affect any of the data sources automations for that same evar. Thanks in advance!
Solved! Go to Solution.
Topics help categorize Community content and increase your ability to discover relevant content.
Views
Replies
Total Likes
It shouldn't impact anything.... Newly added columns should just continue to work as unmapped classifications.
If your existing data should also set those new classifications, then you will need to add new mappings.
The most recently added version of a classification should take precedence... if you don't have a conflicting rule, then the previous rules should still work, treating the new columns as unset. I am not sure, but I think the "overwrite" is no longer an option... I don't have a file handy to test this.
Assuming you are mapping based on SKU, you might have a row for
12234345363564565 - mapped to classification 1, 2 and 3
If 12234345363564565 doesn't need the new classifications 4 and 5, it should continue to map as it always did... if this needs 4 and 5, then I would create a new row mapping 12234345363564565 to all 5 of the classifications...
However, it's always a smart idea to test if there is a chance to break something. Hopefully you have a Dev or QA suite; you can get the classifications replicated there, and import your original file... check the data. Then test the new updated classifications that you want, and confirm that everything is working as expected.
It shouldn't impact anything.... Newly added columns should just continue to work as unmapped classifications.
If your existing data should also set those new classifications, then you will need to add new mappings.
The most recently added version of a classification should take precedence... if you don't have a conflicting rule, then the previous rules should still work, treating the new columns as unset. I am not sure, but I think the "overwrite" is no longer an option... I don't have a file handy to test this.
Assuming you are mapping based on SKU, you might have a row for
12234345363564565 - mapped to classification 1, 2 and 3
If 12234345363564565 doesn't need the new classifications 4 and 5, it should continue to map as it always did... if this needs 4 and 5, then I would create a new row mapping 12234345363564565 to all 5 of the classifications...
However, it's always a smart idea to test if there is a chance to break something. Hopefully you have a Dev or QA suite; you can get the classifications replicated there, and import your original file... check the data. Then test the new updated classifications that you want, and confirm that everything is working as expected.
@Jennifer_Dungan's reply below is spot on.
Just to add, manually importing new classifications shouldn’t break your existing data source automation as long as you’re not overwriting the current classification keys or structure.
Adobe’s classification system is pretty forgiving when it comes to adding additional columns. As long as your manual import:
uses the same key (e.g., eVar value like customer ID),
leaves existing classification columns blank (so they’re untouched),
and doesn’t change the structure of the feed your automated process uses,
…then it should work without interfering with the automated uploads.
Also, worth noting: when you add new classification columns in the UI, they don’t automatically become part of your automated feed, they’ll just remain unmapped unless explicitly included. So your automation will continue updating only the columns it’s configured for, and your manual import will update the new ones.
That said, if possible, doing a quick test in a QA or staging suite is a good safety net before pushing to prod, just to make sure nothing unexpected happens with lookup/mapping behavior.
Hope that helps!
Views
Likes
Replies