I'd argue you could say company with billions of dollars and the prestige to hire world class programmers. Windows Phone is an amazing piece of software, and if we got it 3-4 years earlier, it may have stood a chance.
Microsoft needs to do two things.
1) Give the mobile group vastly more independence to escape the dying desktop business.
2) Make Windows Phone compatible with Android applications. Bribe every major hardware vendor to put out a big new line of Windows Phones to launch with said compatibility. There would no doubt be big tradeoffs in doing this, but it'd be worth it, and would make use of the classic Microsoft strategy of embrace and extend that they seem to have lost an appreciation for after dominating for so long. It's a trojan horse approach superior to their goofy dual booting idea.
I see no reason Microsoft couldn't acquire 20% of the mobile market. Their currently mediocre strategy has delivered some success in Europe, so imagine what a good strategy could deliver. The real question is, can owning 20% of mobile ever be worth the cost if you don't own the hardware (Apple & Samsung)? I don't think so.
One way to leverage this (and pretty sure MS is doing it) is the opposite of what you suggest: to make their smartphones an effective auxiliary device for PCs, beyond just email and exchange and Office. Something like SmartGlass for productivity. Interestingly, there's already work from MSR along these lines, but it seems some way off.
As for Android compatibility, BlueStacks already goes a long way there. If they want to, they could snap them up anytime, but I'm not sure they'll want to capitulate on that front.
I would argue that this "success" is Microsoft's achievement. It happened just because Windows Phone was made by Nokia, which has (had) a good name and market share in Europe. And we know how it turned out for them..