> But in instances where [the new pipeline's applicants are worse than those of the old], then isn't it the case that implementing this change is unfair to the best applicants?
So, to start with, I'm not sure that it would be necessarily unfair, even under the assumption that that would happen. If I were looking for the absolute "best" candidate, I would have to offer the absolute best pay, the absolute best working conditions, and so on. It would obviously narrow down the set to the extent that it could be really challenging to find anyone, let alone someone outside a majority or something like that.
Most of the time, you're not looking for that person. You're looking for someone who's qualified for the job. You can't really even sample the population widely enough to identify which candidates are "best". In that sense, there are no "best" candidates, just those who fit better than the other folks in the rest of the hiring pool.
Maybe put another way, one's sense of "best" might be worth reexamining. Maybe one particular candidate is really highly qualified in a general sense, but in specific comparison to the team they're being considered for, maybe there are enough of that kind of person already on that team. https://staffeng.com/guides/staff-archetypes comes to mind here; you can't have a team that is all Solvers or all Tech Leads or whatever. You can try to reevaluate your definition of "best" such that it enables a more diverse set of perspectives to take part in a team, where you're optimizing for the broadest possible set of e.g. synergistic psychological traits rather than just a specific performance against a specific test set.
But even assuming it were unfair to the teleologically/meritocratically "best" candidates, that that concept isn't flawed:
Don't hire bad candidates.
Keep your standards high enough that people aren't often getting fired. Ensure inflows to your pipeline don't have surprisingly bad success rates. Track down which parts of the pipeline are failing to keep bad candidates from getting to the interview phase or getting hired and be relentless in optimizing them.
That might not sound like novel or DEI-specific advice, and that's because it's not. You should continue to run your hiring pipeline in a way that produces good outcomes. The only difference here is that you have an advantage in finding lots of different kinds of people and that you're providing opportunities to people who often aren't getting them.
So, to start with, I'm not sure that it would be necessarily unfair, even under the assumption that that would happen. If I were looking for the absolute "best" candidate, I would have to offer the absolute best pay, the absolute best working conditions, and so on. It would obviously narrow down the set to the extent that it could be really challenging to find anyone, let alone someone outside a majority or something like that.
Most of the time, you're not looking for that person. You're looking for someone who's qualified for the job. You can't really even sample the population widely enough to identify which candidates are "best". In that sense, there are no "best" candidates, just those who fit better than the other folks in the rest of the hiring pool.
Maybe put another way, one's sense of "best" might be worth reexamining. Maybe one particular candidate is really highly qualified in a general sense, but in specific comparison to the team they're being considered for, maybe there are enough of that kind of person already on that team. https://staffeng.com/guides/staff-archetypes comes to mind here; you can't have a team that is all Solvers or all Tech Leads or whatever. You can try to reevaluate your definition of "best" such that it enables a more diverse set of perspectives to take part in a team, where you're optimizing for the broadest possible set of e.g. synergistic psychological traits rather than just a specific performance against a specific test set.
But even assuming it were unfair to the teleologically/meritocratically "best" candidates, that that concept isn't flawed:
Don't hire bad candidates.
Keep your standards high enough that people aren't often getting fired. Ensure inflows to your pipeline don't have surprisingly bad success rates. Track down which parts of the pipeline are failing to keep bad candidates from getting to the interview phase or getting hired and be relentless in optimizing them.
That might not sound like novel or DEI-specific advice, and that's because it's not. You should continue to run your hiring pipeline in a way that produces good outcomes. The only difference here is that you have an advantage in finding lots of different kinds of people and that you're providing opportunities to people who often aren't getting them.