He was successful because:
1) He got very lucky in the 90's tech boom. Right time, right place, won the lottery.
2) He had enough scientific, engineering, and science fiction education to appreciate and differentiate the 3 edges of:
* "physically possible"
* "possible within the constraints of a project that I can imagine"
* "possible within the constraints of a project that the market can imagine"
and to dream big nerdy dreams rather than to dream specifically of seeing line go up.
3) He was capable of pushing past the last constraint & occasionally the second constraint in a way that was persuasive enough to investors to power a reality distortion field composed of hope for a future that he understood to be technically possible, and convincing them that he was banking on his own success. This is a rare talent, and why tech co-founders have business co-founders. He was always able to put a down payment on these dreams because of #1.
4) Then once he had the money from investors, he put his companies almost entirely in the hands of enthusiastic engineers that were happy to be allowed to contribute to an epic project with actual funding to build the future, rather than spending their lives rotting away writing technical documentation for a defense prime or writing grant proposals or teaching CS101 in a community college or making the 17412nd mobile gacha game. They were given aggressive goals, an aggressive schedule, not a lot of salary, and not a lot of sleep, but free reign to make decisions, move quickly, and break things. The cult of Musk looked less like a religion and more like a fun university extracurricular research project, but for 60 hours a week w/ room & board paid.
Very little of the 'business community' has any special expertise beyond specific social skills, most of them have forced themselves into very-short-timeframe thinking, and they have no idea what's actually possible, when or how to put their faith in the engineers, or what might result. They distrust this entire caste & culture of people, and they perennially shortchange their efforts.
5) Other companies would have outsourced hundreds of parts on these projects primarily to protect management from liability should a project go wrong and the board demand that somebody be ritualistically fired. SpaceX & Tesla preferred to just build their own fabrication shop and hire people to run it, and vertical integration skipped over the prevailing GARGANTUAN market inefficiencies associated with running a complex private-sector supply chain, evidently because there's a top-down disrespect for the sanctity of MBA-type management and for the idea of being accountable to a board of these type of people.