Did this data pull occur where data in same day could have been changed from the initial times each report was pulled? IE if one pulls data for present day 9am Workspace then an hour later you pull it from the other data Warehouse?    

What if you set a direct rule rule on the specific link click and just use it to set an set variable (event number in your case). the the exit link would hopefully just add the extra event in its beacon. The only challenge is ensure set variable fires before exit link i rule fires.

There is no work around in sense of increasing it in same variable.   I copy to a new evar which then has a limit of 512 if memory serves me.   GLTU

I am not aware of any limitation by sheer numbers processed.   I suspect the sheer volume you encounter is just having instances/variations that though they seem to should fall into one classification are somehow being processed as unspecified.   Could there be some logic that when so many are sent ...

First question how are you classifying things.   Using classification rule builder or Uploading files to be classified?

There is also: Return Frequency you may find useful   regards

Hi please see my comment a bit further down in this thread.   https://experienceleaguecommunities.adobe.com/t5/adobe-analytics-questions/running-a-campaign-only-in-us-but-the-report-shows-many-other/m-p/439694#M35950   GLTU

Here is a thread relating to Google ads metics and conversion delats see my comment a bit further below.   https://experienceleaguecommunities.adobe.com/t5/adobe-analytics-questions/delta-between-adobe-analytics-and-google-dcm-floodlight/m-p/429444#M34935   Basically Google ads report metrics can be...

If you know the bot that is being problematic then the solution to try and remove it from crawling is to fix your robots.txt. Your IT can help you with the agent name it sees in its logs.   There you can specify by bot type/name and even restrict pages you do not want crawled. This will not stop all...

As previous you cant directly classify them..    You could try copy Country metric to a new evar then classify it.   GLTU