I can tell you as a former employee what I think it was. Reed was a developer, and in fact made software to make other developers more productive. He still thinks with that mindset internally.
The entire culture at Netflix is built around enabling engineers to do what they think is the best thing. This in turn attracts engineers who like that kind of environment. I would gladly work with any of my former Netflix coworkers again in a heartbeat.
The way this manifests is the mantra "context not control". From Reed on down, management's job is to tell the people that work for them "this is what we want to accomplish as a company/team/group". It is then up to those people to set the agenda.
So I'd say Reed's execution is that he is great at making sure the right people get hired and the culture remains one of high performance, attracting talented engineers who want to work somewhere where making big changes is accepted.
I wasn't there for the ads rollout, but I suspect it went something like Reed saying, "The time has come for us to roll out ads, we need the extra revenue". Then everyone down the line said, "this is how our department will contribute to that". Then the engineers said, "this is what needs to change" and just made it happen and no one got in their way. They have excellent developer tools that enable rapid prototyping and deployment, so while the engineers were building the ad tools, the content team was hiring the best ad exec they could find, who was very interested in the challenge of building an ad organization from scratch, and then everyone probably got out of their way.
I say this because I was there for streaming, and it went very similarly. Reed (and the other execs) agreed that the technology had now caught up to his vision, and it was time to stream over the internet. So the engineers built it while the content team was created to start licensing content. Then some of the engineers said, "we could do this a lot faster and eventually worldwide if we hitch our wagon to this new AWS thing". And so management got out of the way and said "if that's what you think is best, we will support that". And so they started building on AWS, and hired a bunch of people who expertise in building on AWS (that was how I got in) and the rest is history. We kept hiring great engineers and saying "if we do it this way it will be a better experience for the customers" and for us internal teams, our customers were other engineers, so we made the best internal tools we could to enable developers.
> "we could do this a lot faster and eventually worldwide if we hitch our wagon to this new AWS thing"
Standard reminder, this was for all the non-content delivery parts of Netflix. They have always and continue to run their own CDN in-house, and don't use AWS for that part. It also took almost 8 years to fully transition to AWS:
https://ayushhsinghh.medium.com/case-study-how-netflix-is-us...
That article is full of inaccuracies FYI. It took six years, not eight, and 95% of the transition was done in four years.
Also, Netflix didn't always have its own CDN. We used Level3, Limelight, and Akamai to deliver video from 2010 until about 2012/13. And content is still stored in AWS. It's all rendered there and all the original copies are stored there.
You are correct though that Netflix never delivered video bits to users via AWS, because it was cost prohibitive.
The entire culture at Netflix is built around enabling engineers to do what they think is the best thing. This in turn attracts engineers who like that kind of environment. I would gladly work with any of my former Netflix coworkers again in a heartbeat.
The way this manifests is the mantra "context not control". From Reed on down, management's job is to tell the people that work for them "this is what we want to accomplish as a company/team/group". It is then up to those people to set the agenda.
So I'd say Reed's execution is that he is great at making sure the right people get hired and the culture remains one of high performance, attracting talented engineers who want to work somewhere where making big changes is accepted.
I wasn't there for the ads rollout, but I suspect it went something like Reed saying, "The time has come for us to roll out ads, we need the extra revenue". Then everyone down the line said, "this is how our department will contribute to that". Then the engineers said, "this is what needs to change" and just made it happen and no one got in their way. They have excellent developer tools that enable rapid prototyping and deployment, so while the engineers were building the ad tools, the content team was hiring the best ad exec they could find, who was very interested in the challenge of building an ad organization from scratch, and then everyone probably got out of their way.
I say this because I was there for streaming, and it went very similarly. Reed (and the other execs) agreed that the technology had now caught up to his vision, and it was time to stream over the internet. So the engineers built it while the content team was created to start licensing content. Then some of the engineers said, "we could do this a lot faster and eventually worldwide if we hitch our wagon to this new AWS thing". And so management got out of the way and said "if that's what you think is best, we will support that". And so they started building on AWS, and hired a bunch of people who expertise in building on AWS (that was how I got in) and the rest is history. We kept hiring great engineers and saying "if we do it this way it will be a better experience for the customers" and for us internal teams, our customers were other engineers, so we made the best internal tools we could to enable developers.