Include the person covering you while you work on presentation. That effort is part of the process even when it happens at a different bench.

I had recorded my fixture work but not the colleague taking over my other jobs, and our shortest orders seem to create nearly as much setup work as the long ones.

Separate the order types before averaging them. A stable long run might justify a narrowly maintained setup while mixed short work does not, but that distinction needs your observed workload rather than an optimistic label such as production run.

Also include any handling benefit you can actually observe. Time matters, but it need not be the only reason for a bounded application.

Agreed, Hazel, provided a comfortable long-run demonstration does not answer the short-order changeover question. They can remain two different decisions.

The recent-order comparison shows the mixed short work consumes more combined staff time with no clear handling benefit reported; the repeated refill format is the only group still worth a separate trial.

Have you stopped adding presentation parts for the mixed orders, or is the comparison still only a report?

We have stopped those additions and returned mixed orders to the existing method; the repeated-format trial has a capped budget and an owner other than just me, with no claim yet that it earns continued use.