Suggestion for how to handle not having dynamic queries

I'm looking to see if anyone has run into a similar issue and to see what you might have done to combat it.

Dataset 1: Information regarding a file running through a system (FileId, date process, type, file name, etc)

Dataset 2: Information regarding any errors that occured during processing of the above information (FileId, ErrorName,ErrorDetail,DateProcess)

So basically Dataset 2 specifically contains error information as an outcome of Dataset 1 where FileId is a common field between the two.

I have a desire to have a card+drilldowns to be the following

• Top Card: Dataset 1
• Drilldown 1: Dataset 1
• Drilldown 2: Dataset 1
• Drilldown 3: Dataset 2

Now, keep in mind Dataset 2 contains many less columns and really only the FileId field is common between the two, HOWEVER because Dataset 2 doesn't contain most of the information in Dataset 1 when you drill down to it it's obviously completely blank because you're trying to pass filters down to it that it doesn't have a concept of (Ex. Type).

Yes I can use a dataflow to join them together (which I've done) but this isn't very scalable. Once joined a ton of the information in this new Dataset (Dataset 3) is duplicative of both. Ideally we would like some sort of dynamic join when you go to drilldown from Drilldown 2 to 3 but Domo doesn't offer that.

Any suggestions? Besides the data being duplicative it also takes a long time to run due to the # of records (Dataset 1 ~300k, Dataset 2 ~30k, ~2hours to run).