"I had no idea that Amazon's culture was so unique and such an integral part of the company before reading this book. The authors were two early high-level Amazon execs, and they do a great job of explaining the culture and mechanisms that make Amazon unique (first half of the book) and giving examples of Amazon's uniqueness in action (second half of the book).

Chapter 1: Delight the Customer

Takeaways:
1. Customer obsession is KEY. It’s more important than anything else. (Long-term shareholder value is fortunately aligned with pleasing customers).
2. Ensure that financial incentives are aligned with desired results. In particular, you must make sure to not disincentivize collaboration and long-term thinking.
3. Operational excellence is a MUST. There’s no magic formula for this beyond planning well, working hard, and taking responsibility/ownership.
4. Leadership principles must be enforced via mechanisms and baked into every process/aspect of the company.

Chapter 2: Hiring

This chapter is a great "how-to" for creating a decent hiring process.

Takeaways:
1. What makes a good process: it’s simple to understand, can be easily taught to new people, does not depend on scarce resources (such as a single individual), and has a feedback loop to ensure continual improvement.
a. The most interesting aspect to me is the feedback loop for improvement � the Core Bar Raisers train Bar Raisers, who in turn use the debriefing meetings as coaching sessions to engrain the process into interviewers.
2. The 14 Leadership Principles seem like a great way to assess candidates / maintain a corporate culture (as far as that is possible).
3. Hiring is really critical, and really hard, so it makes sense to devote significant resources to it and train a corps of “Bar Raisers� who are voluntary experts in the process.
4. Hiring is complicated, so you need to make sure that you develop a good process, maintain metrics on it, and learn the process really well.

Chapter 3: Organization

Takeaways:
1. Organizing different capabilities into independent units is a must.
2. Building a company is about more than building a product � it’s about building an organization. The organization itself is a type of ‘product� that must be experimented with, iterated on, documented, and improved.

Chapter 4: Communication

Takeaways:
1. You need to make the most of your meetings. They need structure, they need to facilitate efficient information transfer between the presenters and the audience, they need to engage the audience.
2. Narratives help to accomplish these things.
a. Written narratives are much more information dense than slides.
b. The silent reading at the beginning of the meetings prevents interruptions.
c. They allow the efficient communication of complex ideas.
d. They don’t depend on presenters� speaking skills, and their form allows them to be easily distributed and edited.
3. This meeting structure seems good:
a. 1. Silent reading of the narrative (1/3 of meeting)
b. 2. Get feedback from everyone in the room
c. 3. Allow questions/discussion
d. Must record all feedback and questions
4. What do you want to accomplish in a meeting?
a. Force the presenters to think deeply about what they are presenting. Writing a narrative accomplishes this.
b. Enable the audience to understand the idea and to provide useful feedback.
5. Narratives are iterative processes. They are also difficult. Writers must practice writing them, and audience members must practice giving feedback, over many cycles.

Chapter 5: Working Backwards

Takeaways:
1. The PR/FAQ format itself sounds kind of gimmicky, but it’s as good as anything at achieving a great goal � forcing the team to focus on the customer while also enforcing due diligence on the business/product/tech development side.
2. The PR/FAQ format also has the same benefits of the narratives discussed in the previous chapter � it’s a standardized format that’s accessible to anyone in the organization; it’s a complete record of the idea that can be easily passed along, reviewed, and edited; and its written format enables a richer narrative and higher information density than a presentation.
3. There are three parts of the PR/FAQ: the PR (1 page, forces a perspective shift to the customer). The external FAQ (continues to force a customer perspective). The internal FAQ (due diligence on the business/product/tech side).

Chapter 6: Metrics

Takeaways:
1. Running a company is eerily similar to setting up an optimization problem. You need to get the reward function just right, otherwise your optimization problem is going to give you unwanted results. It’s a little more complicated than this though. Although the output metrics are the final ‘reward signal�, since you don’t have direct control over these metrics, you have to define ‘controllable input metrics� which are basically proxy reward signals for your employees. It is a difficult iterative process to make sure that the controllable input metrics are properly aligned with your overall goals.
2. They suggest using the Six Sigma DMAIC (Define-Measure-Analyze-Improve-Control) process for making metrics.
3. The Weekly Business Review meetings are important for evaluating the metrics and using the metrics to detect anomalies/bad trends in the operation of the business. The authors give a lot of guidance on how to structure these meetings to be productive.
4. Looking at anecdotes and exceptions is a crucial complement to looking at the metrics. The anecdotes and exceptions are a reality check that help to expose flaws or oversights in the metrics.

The second half of the book is about how the Kindle, Prime, Prime Video, and AWS were developed. It was pretty cool to hear about these success stories, since they're a big part of what differentiates Amazon from other tech companies (the ability to consistently make successful new products).

Kindle:
The decision to leap into hardware creation for the Kindle was huge for Amazon, but it was carefully planned - they realized that they couldn't differentiate themselves from other e-book sellers based on their selection or price, as they had with physical books, so they had to differentiate via either their content creation process or the reading experience, and they chose the reading experience. Next, they didn't compromise on devising and following through on features that would make the Kindle a truly great reading experience - the E-Ink display, the free cellular internet collection that allowed you to download books from anywhere, etc.

Prime:
The development of Prime was earlier and more haphazard than their other examples, but it's a good example of the Amazon process. They brainstormed a ton of different ideas, most of which were failures or only partial successes, before they finally found the winning formula with Prime. Furthermore, they weren't afraid to pivot even after devoting significant resources to a given strategy. The introduction of Prime guaranteed that Amazon's previous fulfillment strategy (building large fulfillment centers close to distribution centers) would become obsolete (they would need to build many smaller fulfillment centers closer to customers in order to have free two-day shipping), but they went ahead with it full steam anyway. Furthermore, they did it during the busy holiday season when they could easily have claimed they were too busy with other things to create this huge new feature.

Prime Video:
This was the most interesting story for me because it was so full of failure. Amazon's first attempt at video was Amazon Unbox, which was a disaster. Then they made Amazon Video on Demand, which also was not a success. Then they made Amazon Instant Video, which also didn't really go anywhere. Finally, they folded their video service into Prime, which made it a bigger success. Furthermore, they finally decided to go from being just a video aggregator to being a content creator and a device maker. On the device side, this line of thinking led them to develop the Fire Tablet and the Echo. On the content side, they have made a number of hit shows for Prime Video. This example really shows how persistence pays off, yet I think the authors don't highlight the causes of their failures enough. The authors mention that their team was almost entirely MBAs and engineers and no Hollywood people, and some of the key insights came from the rare team members with Hollywood experience - it seems pretty boneheaded to me to not hire more people with relevant industry experience. Also, it took them forever to stop being a video aggregator and instead go after content creation and device making. And finally, it took them forever to realize that the streaming business model was a game changer. They could have saved probably $100m's if they had had some more humility and had spent more time on self-reflection.

AWS:
This chapter was disappointingly short. But AWS clearly shows just how far-sighted Amazon's leadership was. AWS was initially just a service for querying product info for Amazon's referral program, and it eventually become like an API for Amazon's web store. I think it took true insight to realize that instead of just providing an API to Amazon's web store, they could provide an API to Amazon's actual servers and computing power and that this would become an enormous market.

While reading this book, I was getting a lot of Julius Caesar vibes from Jeff Bezos and Amazon. When I was reading Caesar's Gallic Wars and Civil Wars (basically yearly reports of his military activities), I was amazed by how Caesar was able to successfully operate in literally every theater of the Mediterranean in every type of battle in every battle condition against every type of foe. Amazon strikes me as a company with a similar capability - an ability to successfully operate in almost any business. Not only do they have that ability, but they are not afraid to act on it and make big bets. The themes that I picked up on throughout the book were that Amazon succeeds because for them, the company/organization is one of their products, so they are constantly scrutinizing their processes, developing mechanisms to make their success repeatable and to prevent failures, and making long-term investments in both their businesses and their (high-level) employees. Furthermore, if you're really serious about something being part of your company culture, you need to bake it into everyday life at the company. For example, incorporating the 14 Leadership Principles into the hiring process forces the interviewees to use them in evaluating candidates. PowerPoint slides are literally banned for most important meetings. Salary compensation is capped at $160k so that most compensation is in the form of stock. Etc etc. Leadership Principles are corny talking points until you bake them into every aspect and mechanism of the company, and then they suddenly become a way of life.

The one thing that surprises me is that Å·±¦ÓéÀÖ can have such a god-awful UI despite the fact that it's owned by Amazon."