Amazon's lessons on high performance management for the next PM
Being slow is expensive... Planning... Hiring... Incentivising people to seek truth & focus on long-term value, not social pressures... Responsibility/authority... Metrics... The Five Whys
‘If you're good at course correcting, being wrong may be less costly than you think, whereas being slow is going to be expensive for sure.’ Bezos
‘The startups that do things slowly don't do them any better. Just slower.’ Paul Graham, founder of Y-Combinator
‘His bread I eat, his song I sing’. Charlie Munger on the power of incentives
Below are notes on a recent book, Working Backwards, about the management of Amazon by two people who worked with Bezos in senior roles. It has many lessons for other enterprises including politics and government where there is very little focus on elite management and operational excellence.
some basic aspects of Amazon culture
the famous 14 principles (and some comparisons with Westminster)
mechanisms by which Amazon tries to enforce and incentivise these principles
how they do planning and prioritisation
how to incentivise the right behaviour, particularly such that creating long-term value dominates short-term social pressures that easily push people away from trying hard to figure out reality
hiring and training
‘single threaded leadership’ (STL), decentralisation and responsibility/authority
banning PowerPoint and what replaced it
‘Working Backwards’: customer obsession in product development
metrics and the Weekly Business Review
I summarise their main points and add some reflections for those thinking about how a serious government could improve Whitehall including some things I saw, did, or tried to do. At the end I summarise these lessons.
Discussion in Westminster suffers from false dichotomies. People on ‘the right’ who don’t know about great management talk as if ‘bureaucracy is a public sector problem’ that can be cured by ‘making government more like business’. People on the left including many defenders of the status quo say ‘government is not the same as business … it’s simplistic to say cutting bureaucracy is the answer … lessons from great teams aren’t relevant to most of government … we need more money…’
They’re both a bit right and a lot wrong. Whitehall can and should learn from great businesses and great managers, though generally not in the ways ‘free market’ Tory MPs say. And Whitehall is different in critical ways that limit the application of some lessons from business in some narrow ways. But at a more general level, the lessons from great private sector management and the lessons from great public sector case studies are the same. Great businesses can and do learn from projects like Apollo. Governments can and should learn from great businesses. The breakthroughs of ‘systems management’ came in the public sector between the 1940s and 1960s then spread to business then were largely forgotten by western governments (though are much studied in China).
One of many useful things about this book is the way it shows how many problems of ‘bureaucracy’ we see in government are the same or similar to those experienced in Amazon — and Amazon is, by common consent of the great judges of these things (e.g Charlie Munger), one of the very best managed organisations in the world.
Overall the book is an anti-checklist for Westminster in the sense that if you look at all the things they do really well and really value, and you’ve worked around No10/Cabinet office, you’ll say to yourelf ‘tick, tick, tick, government does the opposite, opposite, opposite’. It’s similar to my paper on the ICBM and Apollo projects which also illustrated a set of ideas about brilliant management that are an anti-checklist for Westminster.
Sam Freedman recently wrote a blog about me which touched on a few of these issues. A few people have asked me to respond. He makes some reasonable points but misunderstands what the true problems are in Whitehall, what I think and have done regarding management, and what I think should change in Whitehall. He is much more representative of opinion in SW1 than I am. Few in SW1 can see, never mind understand, the system problems they are caught up in — i.e problems characterised by interdependencies that are very hard or impossible for any individual to fix (lack of management talent, hiring processes, incentives etc). This blog is not an answer to his but the subject matter means that if you’re interested in his blog, then you will see this blog indirectly addresses some of his points. Some time I will write an explicit answer.
Please remember that my comments about Westminster are always subject to the strong caveat that a) there are many exceptional public servants who could make far more money doing something else but make sacrifices for public service, b) a more fundamental problem than the failures of the civil service is that politicians do not care and are not incentivised to care about performance and unless this changes we can only expect the same sort of failures over and over.
When I said this years ago nobody wanted to hear it. But in 2021 we saw that even after a global pandemic costing 150k lives and hundreds of billions, Westminster collectively turned away from facing the reasons for the implosion of core institutions in 2020, No10 embarked on an attempt to rewrite history, most MPs tried to ‘move on’ rather than force honesty about what happened, while the media mainly overwhelmed serious debate with noise. Errors of spring 2020 were repeated more than once killing thousands more. Things that worked well but challenged traditional ways of doing things, such as the Vaccine Taskforce, were dismantled rather than learned from and built on. Rather than have a very fast lessons learned process, the PM and other senior figures have repeated the historical pattern — punt an extremely lengthy inquiry led by lawyers far into the future where it will have little practical effect on how critical decisions are made, just as the Iraq inquiry did not fix the problems of the MOD and Cabinet Office.
If you are thinking about Ukraine, then remember that the same broken central institutions that failed so badly on Iraq, the financial crisis and recovery, covid, fleeing Afghanistan, the petrol shortage, supply chain stresses and so on are in charge of UK’s response on Ukraine/Russia. I strongly recommend you do not trust official statements from No10 or the pundits spoonfed from No10 press office or you will be very surprised about how things turn out.
If you want to examine in detail an organisational culture that is much healthier and higher performance than Whitehall, an organisation that actually lives the culture it advocates, then you will find this book interesting.
Blogs over next few weeks will be: Winning the Next War (Rosen) on military innovation, The Kill Chain on future war, The Power Law (new book on VC which is very good). I’ll also do a reading list (long promised) and another AMA and something on Putin/Ukraine.
Keep reading with a 7-day free trial
Subscribe to Dominic Cummings substack to keep reading this post and get 7 days of free access to the full post archives.