I think that the larger problem is not "too late" so much as the trap of success. Let's assume that corporate growth can be modelled on a sigmoid curve, with s

I think that the larger problem is not "too late" so much as the trap of success. Let's assume that corporate growth can be modelled on a sigmoid curve, with small companies growing exponentially and large companies growing asymtotically in mature markets. There are a couple of really big problems that large companies necessarily face: 1. Real, disruptive innovation is never worth it. Not only do you have the issue that exponential growth in small units will not contribute to exponential growth to the business as a whole but if you disrupt your support structure, you lose on the whole. So, large companies have to be conservative and only acquire disruptive players when they reach a certain size. However this furthers the problem because at that point those players will not continue exponential growth for very long. 2. Large companies are far more reliant on streams of income and partner supports that are subject to disruption than small companies. This again makes it harder to pivot. 3. Large companies have much more inertia than small ones, making pivoting even harder. I don't know if it is "too late" for Microsoft because I don't know what "too late" means. Too late for what? Too late to do what? The company is large and changing remarkably rapidly these days (though I saw the beginnings of change starting when I worked there a decade ago). But the future will certainly be painful for Microsoft. Everyone knows this, even everyone who I know that still works at Microsoft. Is it too late for them to survive? Certainly not. But success on the scale of Microsoft in our market is too often a dead end, and getting out of that dead end is never pleasant.