Those conglomerates developed before the age of computers. They designed their operational structures without the convenience of modern technology, and therefore relied on expanding work forces to grow their company. Labor was cheap, and necessary for growth at the time because humans had to do so much more than they do now.
Facebook is a multinational company with only a few thousand fulltime employees. If GE or Microsoft developed in the same technological context as facebook, they would likely employ far fewer people than they do now. The market increasingly favors agility, and monolithic business models are becoming obsolete (if they are not also decentralized in some capacity, a la Uber).
Tesla can become a powerful multinational player while remaining agile.
The conglomerate structure is kept by choice in the case of Berkshire Hathaway. There's a pretty good outline of the reasoning by Buffett in the 50th anniversary letter.
3M is often used as an example in Strategic Management, particularly as an example of Mintzberg's views (strategies emerge and are not planned). I think the emerged strategy is usually labeled "working with surface material" or something.
I guess one could argue "battery company" is an emergent strategy for Tesla in that sense but I think there's certainly planning to it and it wasn't just discovered. But you can use that "post-branding" that is typical of Mintzberg and I think the author of the blog does a good job.
Facebook is a multinational company with only a few thousand fulltime employees. If GE or Microsoft developed in the same technological context as facebook, they would likely employ far fewer people than they do now. The market increasingly favors agility, and monolithic business models are becoming obsolete (if they are not also decentralized in some capacity, a la Uber).
Tesla can become a powerful multinational player while remaining agile.