How Amazon Does It: Decision Making Inside The World’s Most Daring Digital Company

The true magic in how Amazon built its extraordinary success isn’t technological—it’s organizational, says renowned business advisor and author Ram Charan, with lessons for every company.

The digital age demands high-velocity, high-quality decisions, yet decision-making in traditional companies is hindered by so many hierarchical layers, coordination committees and standing councils that their very survival is at stake. Without algorithms and data at the center of their decision-making, traditional companies cannot compete with today’s digital giants. In 1994, Amazon started creating the greatest revolution in decision-making with one of the oldest cornerstones of success in business: customer obsession. For his new book, The Amazon Management System, world-renowned business advisor, author and speaker Ram Charan drew on publicly available information and interviews with high-level ex- and current Amazon employees to crystalize some of the ideas and insights developed by Jeff Bezos and his team. This excerpt details how CEOs and their senior and middle managers can increase the velocity and quality of their decisions to gain an upper hand against their traditional competitors.

This article was based on The Amazon Management System by Ram Charan.

Amazon’s Jeff Bezos focused on the oldest of old business principles—extraordinary customer obsession to the level that each customer is treated individually—and turned it into the ultimate business purpose of his company. As everyone knows, digitization was the engine that enabled the company to treat each one of its more than 100 million customers as if they were shopping at a corner store in a small town. But the true magic in how Amazon built this system isn’t just technological—it’s organizational. It’s all about making better decisions and making them fast. Here’s how they do it—and how you can, too.

High-Quality, High-Velocity Decision-Making

Five rules can guide you toward high-velocity decision-making:

1. Recognize there are two types of decisions.

To achieve high velocity, Bezos empowers his organization and categorizes all decisions into two types. Type 1 decisions “are consequential and irreversible or nearly irreversible—one-way doors… If you walk through and don’t like what you see on the other side, you can’t get back to where you were before,” Bezos wrote in a 2016 shareholder letter. Such decisions should go through a “heavyweight” process—being made slowly and methodically with great deliberations and consultation—to ensure a high-quality decision. Taking type 1 decisions too lightly is a potentially fatal mistake.

Type 2 decisions refer to those that “are changeable, reversible—they’re two-way doors. If you’ve made a suboptimal Type 2 decision, you don’t have to live with the consequences for that long. You can reopen the door and go back through,” wrote Bezos. These decisions “can and should be made quickly by high-judgment individuals or small groups.” As CEO, you should identify the Type 2 decisions and delegate.

The distinction between two different types of decisions and, thus, two different types of decision-making mechanisms must be crystal clear. It beats bureaucracy, analysis paralysis and improves, over time, people’s judgement in decision-making.

2. Don’t make all decisions by yourself.

No matter how hard-working you and your top team are, there are only 24 hours in a day. As your business continues to grow, if decision-making remains concentrated at the top, sooner or later, you and your top executives will become the biggest impediment to rapid growth.

3. Don’t wait for all the information.

As Bezos wrote in his 2016 shareholder letter, “most decisions should probably be made with somewhere around 70 percent of the information you wish you had. If you wait for 90 percent, in most cases, you’re probably being slow. Plus, either way, you need to be good at quickly recognizing and correcting bad decisions. 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.”

4. Don’t require approvals that are lengthy and must go through large numbers of hierarchical layers.

When multiple functions actually need to be involved in approving a Type 2 decision, you can transform the traditional sequential process into a simultaneous dialogue for high-velocity, high-quality decision-making. For example, Amazon project teams are free to choose between internal services and external vendors. In the traditional sequential selection and approval process described earlier, this could take two to three months and often result in poor quality because distortion and information flow takes place.

Amazon is a continuous invention machine. It has developed the fastest invention process. It involves a clear, written definition of the product to be developed in excruciating detail both in terms of outputs and inputs. The senior people select a leader, and the leader selects an integrated cross-functional team—they call it a separable, single-threaded team—whose full-time job is to deliver the project. They do nothing else. They think and act on the project continuously until it is delivered and operational. I have personally observed putting this methodology in one, large company. They developed a new idea in eight weeks, and they changed the intensity of competition in their industry. This methodology has done wonders for velocity and shortening the cycle time of decision-making.

5. Don’t wait for everyone to agree.

Everyone has experienced postponed decision-making due to one or a few people’s objection or absence. In some cases, in a drive to have consensus, one person can exercise veto power. Sometimes a decision is made and then reopened in the third meeting, which is very frustrating, energy-draining and time-consuming—and the quality of decision, by definition, is poor.

To solve this deadlock, “disagree and commit.” Bezos did that when his team wanted to greenlight an Amazon Original show that he felt should be dropped, responding, “I disagree and commit and hope it becomes the most watched thing we’ve ever made.” Just imagine how long that call would have taken if the team had to educate, persuade and finally earn commitment from him.

After all the facts are considered and all thoughts are expressed, as Bezos wrote in a shareholder letter, “if you have a conviction on a particular direction even though there’s no consensus, it’s helpful to say, ‘Look, I know we disagree on this but will you gamble with me on it? Disagree and commit?’ By the time you’re at this point, no one can know the answer for sure, and you’ll probably get a quick yes.”

This is a two-way, rather than a one-way, approach. Leaders can use it for high-velocity decision-making, and leaders should also be prepared to practice this principle themselves, as Bezos did in greenlighting that Amazon Studios original program.

Controlling for Quality

Now that you’re making high-velocity decisions, how do you make sure they’re high-quality? 

1. Find the best truth.

In traditional companies, due to the inevitable delay, distortion and manipulation as information is relayed through layers from bottom to the top and the fact that data resides in silos, many decisions are made far from the truth, the whole truth and nothing but the truth. 

Rick Dalzell, former Amazon CIO and Bezos’ right-hand man, has written that Bezos “tries to find the best truth all the time.” This may sound self-evident, but it is a big challenge for traditional organizations usually characterized by strict hierarchy, managing by fear and a command-and-control modus operandi. As a devotee of customer service, he seems to have learned from Toyota’s culture of seeking the root cause of any defect. He actually recruited a high-level guy who was an expert at Toyota in making sure that everyone can detect any defect and take action.

At Amazon, each metric is owned by a human being. If the owner finds a significant anomaly, even if it comes from only one customer, the owner is required to search through to the root cause of that anomaly. That means looking all the way back to the beginning of the process through to the actual customer experience.

Bezos does this himself, personally looking at data, seeking out any defects, then asking people to diagnose the root cause, which could be a root cause across a system in the company. For example, when the topic of lengthy hold times for customer calls came up during an executive meeting, he picked up the phone and called customer service himself to demonstrate the discrepancy between what he was being told—that one-minute hold times were typical—and the reality—his hold time on that call was over four minutes.

2. Imagine the possible change.

In addition to the best truth in a static sense, Bezos takes it one step further, i.e., always using future-back perspective, thinking about how things will change going forward. For example, back in 2005, most Amazon executives were against Bezos regarding launching Prime, i.e., $79 membership per year to enjoy two-day shipping for free. The objection was well-founded. Given an $8 logistics cost per order, assuming 20 orders a year on average by each Prime member, shipping would cost $160 a year more than the $79 membership fee. Diego Piacentini, an ex-Apple executive, reported that “…every single financial analysis said we were completely crazy to give two-day shipping for free.”

What gave Bezos the unwavering conviction despite the siege from all?

The key factor was his undying devotion to customer obsession and having the conviction that it would be made fiscally prudent. Bezos took a future-back perspective, asking the obvious but commonly neglected question: How will the shipping cost change?

He believed it would drop. Why? Because when customers spend more, Amazon’s volume will increase, and that increased scale would help Amazon negotiate lower prices from shipping vendors and decrease the amount of fixed-cost allocation of each shipment. In addition, with continuous system upgrades, Amazon’s logistics system would continue to “drive down Amazon’s transportation costs by double-digit percentages each year.” In addition, the frequency of customer interaction will provide more data, and data is equity.

3. Combat group thinking.

Bezos places huge emphasis on fighting conformity, avoiding group thinking and resisting the conventional thinking that achieving harmony is desirable. He expects people to challenge him. He demands a quality discussion with people bringing in new ideas, different perspectives and, even better, disruptive thinking. He “believes that truth springs forth when ideas and perspectives are banged against each other, sometimes violently,” writes Stone.

Amazon’s Leadership Principles state that leaders are “obligated to respectfully challenge decisions when they disagree, even when doing so is uncomfortable or exhausting; they do not compromise for the sake of social cohesion.” Amazon employees well understand their obligation, not only to the company but also to the customer and to the shareholder. Also, as John Rossman, former director of enterprise services at Amazon, writes in The Amazon Way, employees have learned that disagreeing with senior executives is actually beneficial to their careers at Amazon.

I personally observed Ray Dalio, founder and co-chairman of the world’s largest hedge firm, Bridgewater Associates, in his meetings for one full year and saw him very effectively seek opposing viewpoints. Also, he made it easier for people to disagree with him when he often said, “Here is my view, and I could be wrong.” He developed a mobile app that he would use several times in a meeting to get honest input from participants anonymously. Each person in the meeting votes yes or no on a decision and explains why. Every word of the “why” is recorded. Later, all that information is analyzed.   

This process was implemented at the investment firm Matrix Capital. After Matrix founder and CEO David Goel implemented the dialogue process, I observed several sessions where the company’s executive team members logged their candid opinion about whether or not to approve a deal and the underlying detailed logic behind their decisions in the mobile app. The data collected during those sessions was analyzed on a longitudinal basis, which yielded biases analysis, conviction analysis, assumptions analysis and logic analysis, and the process improved the judgment quotient of the whole team.

4. Test through the experiment.

When people have genuine disagreements or when the future is murky, rather than endless debates, fuming arguments and efforts to persuade one another, Bezos opts for an experiment. Brad Stone, author of The Everything Store, vividly captured such a scenario back in 2001. “As usual, Bezos battled his marketing executives. They argued that Amazon had to be on the airwaves to reach new customers. As Amazon’s losses mounted, Bezos’s opposition hardened.”

How to make a decision when two viewpoints are so opposite? Bezos asked the marketing department to run TV commercials in two cities, Minneapolis and Portland, and then measure the additional local purchases they helped generate.

The test went on, the result came back, and the conclusion was clear, “not enough to justify the investment.” Based on this powerful test, Bezos decided not only to cancel all TV advertising, but also to make drastic changes to the marketing department. Today, by any measure, Amazon is the number-one brand in the world—and this was achieved through very little mass advertisement or traditional marketing effort. That is because having a customer experience on an individualized basis that is second to none earns customer admiration for the company.

Experimenting is a natural habit in Amazon. When they do their operating plan, they provide a narrative of learning from the previous year.

5. What if a decision goes wrong?

“We will make bold rather than timid investment decisions where we see a sufficient probability of gaining market leadership advantages,” Bezos stated in a 1997 shareholder letter. “Some of these investments will pay off, others will not, and we will have learned another valuable lesson in either case.”

What’s unique about Amazon’s way of learning? The secret weapon is mid-course adjustment. They ask questions such as what factors should have been considered but did not, what assumptions had been made and why some of them were not reasonable, what critical technology breakthrough was bet on and why it did not happen as expected, and the list could go on. Among all the inputs, the most important one is probably people. Even when you make the right decision in the business sense, if you put the wrong person in charge, the endeavor will probably fail.

Amazon management emphasizes and measures input metrics all the way through. The reason is that if the inputs are right in terms of timing, quality and quantity, the goals will be achieved. This is deep in the culture.

Scaling High-Velocity Decision-Making

In essence, decision-making is about making choices. Tough decisions force us to choose between two goods or weigh two evils. There is no perfect solution, and everything has two sides—the upside and the downside.

In a commencement speech at Princeton University, Bezos asked, “Will you choose a life of ease, or a life of service and adventure?”, “Will you wilt under criticism, or will you follow your convictions?” and “Will you be clear at the expense of others, or will you be kind?”

Of course, Bezos didn’t expect every audience member that day and every future reader of the speech to choose paths like service over ease and clarity over diplomacy, but for people in Amazon, he demands that everyone follow the same principles and methodologies that deliver good decision-making.

Now for the $1 billion question: How to scale high-velocity and high-quality decision-making as your company grows?

1. Crystalize the consistent principles.

In his first shareholder letter back in 1997, Bezos explicitly set forth the decision-making principles of Amazon (see sidebar, right):  Why go through what was surely an onerous and taxing effort? The obvious answer is so shareholders can make informed investment decisions regarding Amazon. It’s also so customers can be more willing to build a trust-based, long-term relationship with Amazon. But, much more importantly, it is for all current and future people working in Amazon, so that every single one of them can clearly understand the decision-making logic and is able to make the right choice when duty calls.

2. Specify the consistent methodology, standard operating procedure (SOP).

On June 9, 2004, we witnessed a brilliant innovation in human management practices. From that day on, Amazon embarked a crusade against PowerPoint presentations and bullet points, and a journey toward what evolved into “Six-Page Narratives.”

If you doubt this methodology’s significance, you are not alone. PowerPoint has become the second language of business. Some companies have a stand-alone functional department devoted entirely to producing PowerPoint presentations.

But Bezos’s was not a capricious whim, but a well-thought-through decision to embrace the value of narrative memos that were often time-intensive endeavors involving multiple iterations. Many Amazonians recalled this practice vividly even after they left the company. In The Amazon Way, Rossman wrote, “I can’t tell you how many of my weekends were consumed by this writing and editing process.”

In my teaching at Northwestern University, I required my students to write in two pages, clearly, concisely and to the point, their case analysis. The purpose was to train their brains to sort out what matters from what does not matter and to write that in a way that a teenager could understand it. And I read each paper and made comments.

Why the huge investment of people’s time and efforts? And why do Amazonians present and past revere it so highly?

As Bezos noted in a 2012 interview with Charlie Rose, “When you have to write your ideas out in complete sentences and complete paragraphs, it forces a deeper clarity of thinking.” Authors are forced to run complete analyses, distinguish between subtle nuances, articulate their logic, prioritize various ideas and take full accountability for specific proposals. There is no wiggle room, no hiding place or no safe haven. Everyone has skin in the game and is held accountable.

Another cultural shock for newcomers to Amazon is that nearly every meeting starts with attendees sitting in silence and reading the narratives for 15-30 minutes. This is because business meeting attendees typically begin interrupting presentations on the very first slide, without an understanding of the full picture the presenter has in mind, and the discussion begins to deteriote.

As Bezos noted in the same interview with Rose, “Executives are very good at interrupting….” So reading memos together is a very effective way to ensure that everyone is well-equipped for a high-quality discussion afterwards. As a result, Amazon’s meetings rarely end without clear decisions or specific actions. As ex-Amazonian Samir Lakhani put it, “Bezos has given all employees a standard SOP for ensuring the basics get done well.” 

3. Reinforce the consistent approach in every decision.

One of Bezos’s most conspicuous traits is consistency. He actually attached the 1997 shareholder letter to every single letter afterwards, probably both as a constant reminder to himself, and as a powerful proof to the customers, shareholders and all employees at Amazon.

In 2010, Bezos noted that “customers who browsed—but didn’t buy—in the lubricant section of Amazon’s sexual-wellness category were receiving personalized emails promoting a variety of gels and other intimacy facilitators.” He “believed the marketing department’s emails caused customers embarrassment and should not have been sent” and called for a meeting, reported Stone.

In the meeting, executives “argued that lubricants were available in grocery stores and drugstores and were not, technically, that embarrassing. They also pointed out that Amazon generated a significant volume of sales with such emails. Bezos didn’t care; no amount of revenue was worth jeopardizing customer trust. It was a revealing—and confirming—moment. He was willing to sacrifice a profitable aspect of his business rather than test Amazon’s bond with its customers.

It is such defining moments that convince others of what is important to you and how you will make decisions in tough situations. As Rossman wrote in The Amazon Way, Amazon’s “principles aren’t slogans printed on wall posters and coffee mugs. They are lived and breathed every day by Amazonians from the CEO on down.” We have cross-checked this with the employees and ex-employees who admire this practice.

I urge readers to reflect on each of the Amazon decision-making mechanisms described here as potential ways to improve your organizations for high-velocity, high-quality decision-making.