This seems like a bad business decision, yet is apparently the cherry-picked highlight of his technical acumen.
This isn't software where we can laugh at Uber building their own messaging service. These are somewhat specialized parts that are made for orgs that have no concern of cost controls. he's absolutely right to question everything as that gives him the edge.
You see something similar with Jobs. From reading his bio, I would say his main skill was ability to sniff out BS. So if he wanted something in a certain way or by a certain time and got push back, he was able to ascert whether it was legitimate or he could just push and raise expectations. And more often than not it worked. The best managers I've had always pushed me and knew when something truly wasn't feasible
Reminds me of the "Bill Gates Excel review" story - https://www.joelonsoftware.com/2006/06/16/my-first-billg-rev...
A key reason SpaceX matters at all is that it can launch things into space at a fraction of what it costs NASA, Russia, China or any other rival to.
It sounds like what Davis got was first quote and must threw it out, if this is the actual story it seems pretty strange that they didn't at least try to work that first-bid number down before discarding the idea of outsourcing the part altogether.
Starship gimbals its Raptors electrically as well. Those actuators aren't probably identical to the Falcon 1 ones, but I would guess that there is a lot of inherited knowledge in them.
For a cleansheet part from a engineer who never built one in his life during the early days of this approach to aerospace, 9 months seem reasonable to me.
There's multiple issues here that apparently others aren't seeing.
The key one: it's a build or buy decision for a non-key part.
SpaceX's USP (as far as I'm aware, not a rocket nerd) was reusable rockets. If they can successfully demonstrate that, then they've won. They don't want to run out of money before then but chasing down every inefficiency in the existing system seems more like OCD than good business sense.
Subsidiary issues:
Sending the guy who "had never built anything" to negotiate hardware orders.
Getting the guy who "had never built anything" to build a vital part when off the shelf alternatives were available.
Celebrating this as business genius not unnecessary risk.
And this story, as far as I remember from the book, was from the Falcon 1 days, where reusability was not on the table. Early days type stuff where cash reserve preservation was tantamount.
The guy “had never built anything like the actuator” but he was an engineer who (hopefully by virtue of being an engineer) understood the first principles behind designs.
To put into context, a large part of SpaceX’s unique competitive advantage comes actions like this. The Aerospace industry up until SpaceX mostly operated in the manner you suggest: buy the outsourced parts, design only what is “truly key”, and then cobble it all together.
Except that exposes yourself to large amounts of supply chain risk, design constraints, and price gouging like no other. The actuator costed $120k because there were no competitors and they knew the old primes would buy it anyways.
Elon definitely went against the grain of established thinking at the time and it’s clearly paid off. Maybe not “business genius” but it’s not a small undertaking and went against decades of ingrained engineering culture and business practice.
An off the shelf actuator could mean they would have to adapt other hardware to the actuator. By in-housing their own, they can make the actuator specific to their needs.
Rockets are tricky because of how very specific everything can be to the specific rocket and system that you built.