They have to maintain backwards compatibility for 40+ year old applications so that they don’t lose big corporate and government customers, but they also have to chase the newest trends in order to keep their shareholders happy. They built their business on selling their software, but most of their competitors are giving functionally-equivalent programs away for free. Their software runs without incident on literally billions of devices for decades, but one or two high-visibility bugs or design missteps and public perception of their brand totally tanks.

And so, their business model sucks. Moving Windows to become a data-harvesting SaaS was a terrible choice, their pivot to AI is going to crash and burn, and rent seeking software subscriptions are a scourge.

But I think they’re just too big and too vertically integrated to actually be any better at this point. I just don’t think it’s possible for their executive team to make good decisions anymore, not because they’re dumb, but because the good decisions literally don’t exist. It’s like a black hole, where the closer you get to the event horizon, the more possible paths point toward the singularity; likewise, the bigger Microsoft gets, the more possible decisions point toward “devastatingly bad.” They honestly should have been split up 25 years ago; for the industry’s sake and for their own.

Moving away from Office and Windows and so forth is a nightmare for any larger company. If you use specialized software, it might very well only run on Windows or only have an integration into Office. Even if you could, you then have to retrain staff to use Libre Office, Linux and other alternatives. You also will have problems converting, changing servers and so forth.

So companies just do not switch. That is how Microsoft makes money. They really do not care that much about private users. That is only usefull so people can use their products.