No Rules Rules - Netflix and the Culture of Reinvention by Reed Hastings (Netflix co-founder & CEO) and Erin Meyer
What I like about the book is the story behind the way of working and the current ‘rules’/practices of Netflix. The book shows how Hastings manages the company as a complex system: providing enabling constraints (not too many), trying it out, see where it fails, and adjusting the constraints.
The things I don’t like are woven in further down.
Hastings took his experience from his previous company, where he’d establish new rules whenever something didn’t work out, and set out at Netflix to do this differently. In his previous company he took the road of establishing more and more governing constraints and rules, stifling the company. Managing it as a complicated system, instead of a complex system.
Principles for Netflix:
- Build up talent density - if you focus on high performers, you can get rid of many controls meant for dealing with sloppy, unprofessional, irresponsible behaviour
- Increase candor - employees can learn a lot from each other, you need to get into a feedback habit
- Reduce controls - “lead with context, not with controls”
Netflix wanted a culture of freedom and responsibility
So we should copy Netflix to be successful?
The book describes the current way of working and how they arrived at this through experimenting and adjusting. The risk of book like this is that people take the current state and try to copy it. Copying the motions of Netflix or some other successful company won’t help you.
A complex system like an organization has history. The past/road followed led to the current state of affairs. Copying the motions won’t get you far, as you are confusing correlation for causation. Netflix culture is like it is now because of its history of continuously evolving constraints through experimentation. This is what is helping it to be a successful company. The current way of working is a result of that, not the cause.
This is similar to what happened to the ‘Spotify model’: Henrik Kniberg shared how Spotify was organizing for scaling up while trying to keep agile. The article described a snapshot in time, as Spotify is continuously evolving and adapting to change. Unfortunately many organizations didn’t read Henrik’s disclaimer in the article and starting copying the motions.
Willem wrote about this pattern of copying others without thinking and adaptation in Cargo cults and angry chaos monkeys
Hiring “Top Performers”
Netflix focuses on hiring “top performers”. Prevent employees that pull down morale, by not being good enough, by slacking or being a jerk (no matter how brilliant a person seems to be). Netflix also pays employees above market. No performance-based bonuses
So they hire top performers, provide a vision and context, and some enabling constraints and expect them to do their thing. Treat them as the adult they are. And they say that “Performance is contagious” - focusing on high performing people reinforces itself.
I noticed that Reed is referring to some research about software programmer performance (the “rock star principle”), where high performers outperform low performers by an order of magnitude. Productivity in software development is a not well understood topic, with lots of bad or meaningless research around. Does anyone remember counting lines of code as a measure of productivity?. Some still believe in that to this day. The research cited in the book focuses very much on individual performance, while software development is a team sport.
The performance of a high performing team is more than the sum of the performance of its members. I’m not saying Netflix is taking a wrong approach, their results until recently show the opposite, but what ‘top performer’ means does not become clear from the book, I’d be interested to learn more, especially for software development.
Performance of an individual is is always a function of the individual’s capabilities and the context the person is working in. Although the book mentions they are collaboration focused, I did not see this specific thing explicitly.
Retention: the Keeper Test
An organization is not a family; a family stays together regardless of “performance”.
Netflix asks both managers and employees to frequently do the Keeper Test: “If a person on your team were to quit tomorrow, would you try to change their mind? Or would you accept their resignation, perhaps with a little relief? If the latter, you should give them a severance package now, and look for a star, someone you would fight to keep.”
They focus on “top performance”. If someone shows adequate performance, they will let go of this person with a generous severance - again above market and above what is required by law. So they take a proactive approach to employees that are not a good fit. Compare this to other companies that start some painful improvement program, and just wait until the employee leaves by themselves, or companies that would bully the employee away. Note that attrition is similar to market averages so it is not working out worse for them.
Frequent candid feedback has high leverage: it “exponentially magnifies the speed and effectiveness of your team or workforce”. So build feedback in and coach employees to give/receive feedback effectively - “feedback anywhere, anytime”.
4A Feedback Guidelines:
- Aim to Assist - give constructive feedback, with positive intent
- Actionable - what can one do differently?
- Appreciate - instead of providing defense or excuses
- Accept or Discard - you’re not required to follow it
Netflix experimented with 360 feedback. What they learned:
- do not do them anonymously
- don’t tie the feedback to performance reviews
- live 360s, in a small group, e.g. over dinner; this can take hours and can feel uncomfortable, but it is effective; use start/stop/continue
No holiday policies; no travel and expense policies. Instead of ‘pessimistic control’ (always asking for permission), they apply ‘optimistic control’, finance checks samples after the fact. If people overspend, this does not mean a reprimand, but it triggers to have another look at context.
No policies can also mean that people do not take holiday, because they feel pressured to work. In this context the ‘Keeper test’ and ‘Top performer’ focus may be harmful to the employees health and the company’s performance, as tired people make more mistakes, and burn out more easily.
Trust leads to trustworthiness
Lead with context. Prerequisites for this are:
- a high talent density
- the goal is innovation, not error prevention, like in a context where safety is critical
- the system is loosely coupled, with high alignment
If someone does something dumb, don’t blame that person, but ask yourself what context did you fail to set. I like this systemic perspective, it is in line with Deming’s statement that 90-95% of performance is governed by the system. Contrast with imposing more rules creates an inflexible context where people cannot thrive. This highlights a main job of management: working on the system, creating a context where good results emerge.
I found this book worthwhile to read. Seen through a complex adaptive systems lens, it tells a nice story of the path Netflix followed, what they tried out and why they decided things.
It left me with questions about ‘hiring top performers’, especially in my trade (software development) things like ‘top performer’ and productivity are complex to grasp.
Like with many other books and articles it bears the risk of people copy-pasting the current Netflix practices. If you want to take things from the book to your own organization, focus on the principles and prepare for a journey of trying things out, finding it doesn’t work and tweaking it until it does, with a preference for setting context and trusting people to fix things.
Credits: thanks to Willem for editing and helping to get this post published.