That's the issue, right? Everyone wants "diversity", but no one wants quotas or hard lines that could discriminate against other groups. I think you need to do your best to root out any biases in your hiring process or people. At Uber's size there is also an evangelism role that can be played to reach out and encourage a more diverse applicant pool to apply for roles. But this report falls flat on how you could possibly measure whether your team is "diverse" or "inclusive" enough and also fails to call out any existing practices, processes or people that would've gotten them there in the first place, opting instead to use those principles as PR tools.
I don't know things like "we only bought jackets for the men, because it would be more expensive to buy jackets to the women... so they just don't get jackets" seems pretty divisive to me. You're trying to split hairs, but things like that are so blatant that they are screaming right in your face. If such things were really happening at Uber, then I don't think that your quibbling about "how diverse is diverse" makes any sense. They obviously have issues that need to be addressed, no?
There's the aphorism that what's measured gets improved (and what isn't measured gets worse). If you're trying to improve diversity/inclusion, how do you know you've done a good job unless you can measure it?
It also depends on how and what you're measuring. If you are (e.g.) measuring diversity, but measuring it in the "wrong way" you could be setting up perverse incentives to game the metrics while not truly achieving the goal.
I feel like the rule should be fairly simple: your organization should match the ethnic and gender breakdown of the area it's offices are in.
So, for example, if Silicon Valley is 10% African Americans (I have no idea if that # if true, just using it for argument's sake), then your organization should be 10% African American. This also means your organization should be 50% women, since they are ~50% of the population. How you manage that (blind hiring, quota hiring, etc) is up to the company, and is difficult. But measuring proper diversity is really easy: you should reflect the local city's diversity. If you don't, you need to have a really good reason (and "we don't think we can be drinking buddies with them" is a really shitty reason).
Can you control for education? Why is it a company's responsibility to counteract society's bias? Even if one company could succeed at that excellent goal, it'd be impossible for all companies in an area to follow that rule. Unless many people were unemployed.
You don't "measure" zero tolerance. And there's no success here--that is, there's no point at which you say "this problem is solved." You commit to throwing out the bad apples, you throw out the bad apples, and promise to throw out any bad apples as you find them in the future.
Claiming zero tolerance is a desire to stay ignorant of your measuring precision. You need to know how accurate your tools are before you can estimate how many incidents occur.
If you pick a "zero" tolerance, that's saying your measurement can actually be zero and there's no need to improve from that point. If you pick a measurement that can only asymptotically approach zero, then you are recognizing that improvement is always possible.
But if you're able to measure a problem, that means you've identified a violation of your zero tolerance policy. Which begs the question, why is your measurement more effective at finding the problem than the enforcers of the policy?
If you're not half-assing your zero tolerance policy, measuring it is immaterial. Find an offender, fire them. The number of firings in hindsight isn't a factor towards your future, since those people are gone. There's not a meaningful measurement that you can make, because after you fire someone, they don't impact your numbers anymore. You assume the problem is ongoing and will continue to be ongoing, you immediately and harshly deal with the problem when you encounter it, and you move on to finding more problems. There is no gray area.
> after you fire someone, they don't impact your numbers anymore
They would if your measurement is incidences per month. Otherwise you could fire someone every day and think you're doing a good job at office culture. I'd rather not have any harassment. Of course, you'd need to be careful not to discourage reporting of incidences. Measuring things is tricky.
> Find an offender, fire them
What's your technique of "finding" problems? Could it be improved? How would you know?
> no gray area
Of course there is. How do you know when someone accidentally insinuated sex versus intended a salacious proposal? Or accidentally touched someone's rear while passing by? Because it happened 6 times in 2 days? What about 2 times in 6 months?