You have to consider this in the context at the time. Yes it would have been expensive, but Microsoft was investing BILLIONS into Windows Phone. Microsoft and partners spent something like $700 million dollars just on marketing for the launch of Windows Phone 7[1,2].

To spend that kind of money on marketing and then not dedicate resources to the actual product seems foolish. And I am not saying they should have done this for only one app. I am saying they should have done this for many apps. If they had created quality versions of, say, the top 25 apps for mobile at the time they would have been in a much better position. I believe they could have made significant traction with business users. Remember, at the time Office wasn't available on other platforms and was (is) a huge draw for many people.

If they had been successful with the strategy and gained market share the partners would have wanted to take over their own apps anyway to enable monetization. But they needed users for that and to get users they needed apps. You have to jump start it somehow.

Now, would it have made any difference? Who knows. But IMO, you either need to not do it or you need to do all parts of it right. You can't go half way on the ecosystem and expect to succeed in an already challenging market.

1: https://techcrunch.com/2010/08/26/microsoft-half-billion-dol... 2: https://techcrunch.com/2012/01/04/microsoft-oems-pledging-20...