Comment by Puzzled_Cheetah

In the short run, I'd mostly agree with you. Something being parallelisable at a low cost lets you do more of it in a short space of time, and probably lends itself to what should be an earlier decision as to whether to reexamine your approach. You're going to have a lot more information earlier - all else being equal.

In the mid to long run, however, the costs mount up - however small they may start off as. You can make, say, 20 job applications a day - (not an uncommon thing to have people do if they were on mandated provision) - have a couple of hundred open in a few weeks and get nothing back. That takes something out of people. People will still spend time from finding and making those applications, energy, motivation, an awareness of the wider context in which their problem's structured....

There's a human component to the problem which generates opportunity costs for things that don't seem to have any. Like yes, technically, you can put 20 job applications in a day and then go and volunteer. But it's the rare person who's actually going to have the energy, motivation and awareness to do that when they're going hard on a broad search strategy.