Because customers do not know they have raised a ticket. They email, get an auto-reply, worry, use the chat widget, then ring — and three agents start work on the same problem.
Merging combines tickets about the same problem into one, keeping every message and both histories.
Tickets merge into a chosen primary, keeping all correspondence in time order. The merged ticket keeps pointers, so a customer following an old reference still lands in the right place.
A customer emails on Monday and uses the widget on Tuesday about the same fault. Merging gives one conversation, one owner and one clock, with both threads visible.
It will not un-merge cleanly if you merge the wrong ones. Splitting is possible but the conversation has already been interleaved, so it is worth checking first.
The clock follows the earliest ticket, so merging cannot be used to reset an SLA, and reporting counts one resolution rather than three.
They get one thread. Old references still resolve.
The earliest, so merging never buys time.
No. That would cross a privacy boundary.
Fourteen days, every module, no card. Or half an hour with someone who will run it on your own records and tell you where it does not help.