There's a pipeline problem if you are only taking a specific output and society is not producing it. As an individual company then you can either broaden what you accept (and I don't think this will lower quality, but it will make hiring harder), or take charge and extend the pipeline in your own backyard through training or probationary periods to trial someone.
FTA: > As you can see, broadening your pipeline isn’t a magic pill, and as long as demand for software engineers continues to grow, it’s still going to be really hard, systemic changes to our society notwithstanding. If we do make these changes, however, the tech industry as a whole can accelerate its path toward gender parity and potentially get there within a decade.
There's probably a feedback loop here where if they can broaden their intake, it will help accelerate societal change. The largest companies will have the most trouble, because they have the most seats to fill, but they also have the resources to try something different to fix this if they really want to.
On your last point, I think having a hard target like 50% can be damaging since it is demoralising to look at the gap and it might all feel like too much to deal with. I'd personally prefer to work towards 'a bit better than last year' with some achievable (non-startup) growth rates.
I'm not saying a company should throw up its hands and do nothing. But why are we looking to assign to companies the solving of a 50% problem (if that's the goal), when they can only ever achieve 15% within their reasonable powers?