Book Review: Working Backwards


Working Backwards: Insights, Stories, and Secrets from Inside Amazon was recommended to me by Brian Gibson. It’s an easy read, with some interesting details about the history of Amazon.

Amazon is customer obsessed, and Jeff Bezos is quoted as saying that customer value is completely aligned with shareholder value. As such, making decisions in the customers’ best interests will yield dividends for shareholders, whereas too many companies today focus more on shareholder value and produce sub-standard customer experiences and products. Amazon thinks long-term, and a phrase that pops up throughout the book is that the company is willing to be misunderstood for long periods of time because they’re certain that their long-term plans will produce the results they desire. This is different from the quarterly profits chased by so much of Corporate America today.

The book opens with the 14 Amazon Leadership Principles:

Each of these has some level of explanatory detail provided, and they are each referred to throughout the examples provided in the book. Some history of how these were developed is provided, and with this baseline the book then digs into several specific things that Amazon does that the authors feel are sufficiently unique and valuable.

The second chapter is on the hiring process at Amazon.

When you consider the potential positive and negative impacts of an important hire, not to mention the precious time dedicated to it, it is shocking to consider how little rigor and analysis most companies put into their hiring process.

I can attest to this through my own experiences. I think the general trend is improving, but there remains an enormous amount of opportunity for improvement at most companies. The book shares some of the missteps Amazon had in their hiring process, due in no small part to their rapid growth.

Another force that works against successful hiring is the lack of a formal process and training. Startups and rapidly growing companies are particularly likely to hire new people without a process in place, though all too often many more-established companies have the same problem.

The process Amazon developed is an interesting one, and it’s worth examing what they do and why they do it.

The third chapter is about organizing humans to solve problems. There is a history of the evolution of the “two pizza” teams that Amazon pioneered, as well as some examination of what it means for a team to be empowered and autonomous. This is hard stuff, and takes some real committment from management to make work well.

One of the interesting realizations that Amazon had was that they didn’t want to improve the communication between teams in order to get more work done. They wanted teams to communicate less! That’s not to say that they didn’t want human beings to socialize or collaborate, but if my team needs to coordinate with your team to get something done, then my team can only be as effective as your team can facilitate. That kind of interaction can slow everything down, and it was these sorts of communications that Amazon sought to eliminate.

And we came to realize that other limits to autonomy would also need to remain, with each team still tied to others by varying levels of dependency. While each two-pizza team crafted its own product vision and development roadmap, unavoidable dependencies could arise in the form of cross-functional projects or top-down initiatives that spanned multiple teams. For example, a two-pizza team working on picking algorithms for the fulfillment centers might also be called upon to add support for robotics being implemented to move products around the warehouse.

We found it helpful to think of such cross-functional projects as a kind of tax, a payment one team had to make in support of the overall forward progress of the company. We tried to minimize such intrusions but not avoid them altogether. Some teams, through no fault of their own, found themselves in a higher tax bracket than others.

With respect to management, this chapter details the “single threaded leader” concept, where a person is solely responsible for one thing. If a manager or leader is responsible for one thing primarily, and one or more ancillary things, it’s no surprise that the ancillary things will get less attention and are less likely to succeed. Amazon solves this by dedicating people to one thing at a time.

Chapters four and five had, for me, the most meat, dealing with communication and working backwards, respectively. Amazon does not use Powerpoint or similar presentations when communicating new and complex things. Instead, they use a long-form “six pager” document of, roughly, six pages. This document explains things in detail, with supporting information and an attempt to address concerns and objections up front.

These documents are distributed at the beginning of meetings, and attendees spend the first part of the meeting reading them. Then, and only then, is the topic discussed verbally. This has several interesting and non-obvious benefits. Because the document is distributed at the beginning of the meeting, you have literally until the moments before the meeting to finish writing, which may be a generous more amount of time than might be required if you have to submit your “deck” the day before the meeting. The silent reading period ensures that the document is fresh in everyone’s mind, and they have the chance to jot down notes as they need.

The inclusion of counter-arguments and concerns in the documents forces the authors to critically evaluate their proposals in advance, to demonstrate to their peers and leaders that this isn’t some “ponies and rainbows” fantasy proposal.

I’d read about the six pager previously, and even tried my hand at writing a few at work. It’s hard, and time consuming, and frustrating. But I think it’s a worthwhile exercise. In my cases, the documents were not read at the beginning of meetings, so I only got half of the potential benefit, but the exercise of writing down, in detailed prose, my proposals was very helpful. I recommend that all teams give this approach a real try for a couple of iterations, to see what benefits it provides them.

The titular “working backwards” section was also interesting. Amazon encourages teams to write press releases for products they haven’t even built yet. These press releases are for internal distribution only, and like the six pagers they attempt to capture a lot of detail about something. These documents help the teams identify what the ideal end state is, which informs many decisions they’ll need to achieve that state. Several examples are provided, and it’s an interesting approach.

The last half of the book is a selection of longer examples of all of the previous material as it applies to specific Amazon products: Kindle, Prime, Prime Video, and AWS. The “behind the curtain” view is interesting, and the authors do candidly share some missteps and mistakes along the way, but on the whole these examples weren’t super helpful to me. They’re proof that the Amazon process works, sure, but they each span several years of efforts, told in a not-entirely linear fashion, and they gloss over an awful lot of hard work from a lot of people.

That’s one of my biggest complaints about the book, actually: it focuses on “leaders” and managers and barely discusses the massive contributions of the many individual contributors who did much of the real work behind all of these products.

There’s also a lot of bias in the selection of the stories provided. There is a passing mention of the Amazon Fire Phone as a dud, which was disappointing. In what ways did the execution of all of these Amazon principles and experiences conspire to produce a dud? How did the working backwards process fail to identify what customers really wanted? There’s a good bit in the book about Amazon’s attitude toward failure, and how it’s a learning opportunity. I’d’ve loved to learn more from the Fire Phone failure than from the Prime Video success.

All in all, this is a decent book to read. There’s some really thought provoking stuff for people at companies of all sizes, and there’s a fair bit of stuff that seems best for giant companies with plenty of capital reserves. Not every company is as willing to take as long-term a view as Amazon, which is a shame: if more companies looked beyond quarterly earnings they could accomplish a lot more.

home / about / archive / RSS