Customer Obsession is Amazon’s #1 Leadership Principle. Everyone knows it. Everyone studies it. Almost everyone gets it wrong in interviews. After spending a few years as a Senior TPM at Amazon and conducting 100+ interviews, I’m going to tell you what Customer Obsession actually means from the inside. And it’s not what you think.
The Answer That Gets You Rejected
Let me paint a picture for you.
I’m sitting in an interview room. The candidate across from me has 10 years of experience. Strong resume. Impressive companies. I ask them: “Tell me about a time you were obsessed with the customer.”
And they say this:
“A customer reported that our dashboard was loading slowly. I immediately prioritized the fix, worked with the engineering team over the weekend, and we shipped a performance improvement that reduced load time by 60%. The customer was really happy.”
Sounds great, right? Customer had a problem. They fixed it fast. Customer was happy.
That candidate did not get an offer.
And here’s the thing: if you’re reading that story and thinking “Wait, what’s wrong with that?” then this article is going to change how you think about Customer Obsession entirely.
Because that story? That’s customer service. That’s responding to a complaint and fixing it. That’s table stakes. That’s what any decent company does.
That is not Customer Obsession.
Let me tell you what is.
The Three Dimensions of Customer Obsession (That Nobody Talks About)
After years of living inside Amazon’s culture, conducting interviews, sitting through debriefs, and watching who gets hired and who doesn’t, I’ve come to understand that Customer Obsession operates on three distinct dimensions. These aren’t written in any Amazon document. They’re not in any interview prep guide. But they are, in my experience, exactly how Amazon evaluates whether you truly embody this principle.
And the reason most candidates fail? They only understand the first dimension.
Dimension 1: Work Backwards from the Customer
This is the one everyone knows. Amazon starts from the customer and works backwards. The PR/FAQ process. You write the press release first, the one that describes what the product or feature will do for the customer, and then you build the thing.
Most candidates understand this conceptually. They’ve read about it. They can talk about it.
But here’s what they miss: Working backwards isn’t just a product development process. It’s a thinking discipline.
It means every decision, whether technical, prioritization, or resource allocation, starts with one question: “What does the customer need?”
Not “What can we build?” Not “What does our roadmap say?” Not “What does our VP want?”
What. Does. The. Customer. Need.
I remember a planning meeting at Amazon where a team presented a beautiful technical architecture. Elegant design. Clean abstractions. The senior leader asked one question: “Which customer asked for this?”
Silence.
The team had built something they thought was cool. They hadn’t started from a customer need. The project was killed in that meeting.
That’s Dimension 1. Start from the customer. Always.
How to show this in an interview:
“Before building anything, I talked to 15 customers to understand what they actually needed. Their stated request was X, but through those conversations I discovered the real pain point was Y. So we wrote a PR/FAQ around solving Y and worked backwards from there.”
Dimension 2: Anticipate What the Customer Can’t Articulate
This is where it gets interesting. And this is where most candidates completely fall short.
Dimension 1 is about listening to customers. Dimension 2 is about thinking ahead of customers. It’s about understanding a customer’s pain so deeply that you solve problems they don’t even know they have yet.
Think about Amazon Go and Just Walk Out technology.
No customer walked into a grocery store and said: “I wish I could just grab things and leave without checking out.” Customers weren’t asking for that. They were used to checkout lines. They accepted them as a fact of life.
But Amazon observed the pain. They saw people abandoning baskets when lines were long. They saw the friction. They understood, at a deep level, that the checkout process was a barrier between the customer and what they actually wanted: getting their groceries and leaving.
So they eliminated it entirely.
That’s not customer service. That’s not even working backwards from a stated need. That’s obsessing over the customer experience so deeply that you see friction the customer has learned to tolerate.
Henry Ford supposedly said: “If I had asked people what they wanted, they would have said faster horses.” Whether or not he actually said it, the principle is real. Customers can tell you their pain. They usually can’t tell you the solution.
I saw this play out at Amazon constantly. The best product decisions weren’t responses to customer requests. They were responses to customer behaviors. To patterns in the data. To friction that customers had normalized.
How to show this in an interview:
“Customers weren’t complaining about the onboarding flow. Our satisfaction scores were fine. But I noticed in the data that 40% of new users dropped off at step 3. They’d accepted the friction. So I redesigned the flow to eliminate that step entirely. Completion rates jumped 35% and we saw a measurable increase in 30-day retention. Nobody asked for it. The data told us.”
This is the dimension that makes interviewers lean forward. Because it shows you’re not just reactive. You’re proactive. You’re not waiting for customers to tell you what’s wrong. You’re finding it yourself.
Dimension 3: Cannibalize Yourself for the Customer
Okay. This is the one that separates Amazonians from everyone else.
And honestly, it’s the one that most people outside of Amazon find genuinely uncomfortable.
True Customer Obsession means being willing to hurt your own metrics, your own team’s roadmap, your own short-term P&L, if that’s the right thing for the customer.
This isn’t theoretical. This is Amazon’s actual history:
| Decision | Short-Term Pain | Long-Term Payoff |
|---|---|---|
| Amazon Marketplace: Invited competitors to sell on Amazon.com | Cannibalized first-party retail revenue | 60%+ of Amazon sales are now third-party sellers |
| AWS price cuts: Over 100 price reductions, often before customers asked | Left billions in revenue on the table | Made it nearly impossible for competitors to catch up |
| Kindle hardware sold at cost (sometimes at a loss) | Zero hardware margin | Built the largest ebook ecosystem on earth |
| Prime at $79/year when shipping costs far exceeded that | Massive per-member shipping losses | 200M+ members spending 2-3x more than non-members |
Think about that Marketplace decision for a second. Jeff Bezos invited competitors to sell on Amazon’s own platform. His retail team was furious. They were literally inviting other companies to take their sales.
But Bezos understood something: customers wanted more selection and lower prices. If Amazon couldn’t offer that through first-party retail alone, then the customer-obsessed thing to do was to let third parties compete, even if it meant Amazon’s own retail business would take a hit.
That’s not customer service. That’s not working backwards from a feature request. That’s being willing to cannibalize yourself because it’s right for the customer.
And here’s Bezos’s famous line that captures this perfectly:
“Your margin is my opportunity.”
Translation: If you’re comfortable with how much value you’re keeping versus how much you’re giving to the customer, you’re not customer obsessed enough.
Now, I know what you’re thinking. “This sounds like a charity, not a business.” Hold that thought. I’m going to address it. But first, let me tell you why this dimension matters in interviews.
How to show this in an interview:
“We had pressure to ship Feature X. It was already on the roadmap, engineering was ready, and it would have boosted our quarterly metrics. But I discovered through customer research that what users actually needed was a fix to an existing workflow that was causing them daily friction. I made the case to leadership to delay Feature X by three weeks to solve the existing pain point first. It hurt our sprint velocity numbers, but customer satisfaction scores for that workflow went from 3.2 to 4.6, and churn in that segment dropped 25%.”
That story is powerful because it shows you were willing to sacrifice something for the customer: your team’s roadmap, your velocity metrics, your quarterly goals. That’s the signal interviewers are looking for.
Compare that to what most candidates say:
“A customer had a problem, so I fixed it.”
That’s customer service. It’s expected. It’s unremarkable. It tells me nothing about whether you’d make the hard call when customer needs conflict with business convenience.
The Three Dimensions Together
Let me give you the complete picture. Here’s how these three dimensions stack:
| # | Dimension | One-Liner | Signal to Interviewer |
|---|---|---|---|
| 1 | Work Backwards | Start from the customer need, then build | “This person starts with why, not what” |
| 2 | Anticipate the Unarticulated | Solve pain the customer can’t even name yet | “This person thinks deeper than the surface” |
| 3 | Cannibalize Yourself | Sacrifice your own metrics when it’s right for the customer | “This person makes the hard calls” |
Most candidates only show Dimension 1. Good candidates show Dimensions 1 and 2. The candidates who get offers at senior levels? They show all three.
And here’s the thing: you don’t need three separate stories. The best Customer Obsession stories weave all three dimensions into one narrative. You started from the customer (Dimension 1), you discovered something they didn’t even know was a problem (Dimension 2), and you sacrificed something to solve it (Dimension 3).
That’s a story that gets you hired.
“But Wait. Amazon Is a Business, Not a Charity”
If you’ve been reading this and thinking: “This all sounds great, but at some point you have to make money, right?” Good. That’s a smart question. And honestly, it’s one that comes up in interviews more than you’d think.
Here’s the uncomfortable truth that I had to learn myself:
Customer Obsession at Amazon is deliberately unbalanced. It doesn’t have an equal counterweight. When it conflicts with other priorities, it wins. Always.
I know that sounds extreme. But think about it. In my LP Pairs article, I explained how most Leadership Principles work in balanced pairs: Bias for Action pairs with Insist on High Standards. Ownership pairs with Deliver Results. Think Big pairs with Insist on High Standards.
But Customer Obsession? It’s LP #1 for a reason. It doesn’t have an equal and opposite principle that says “don’t be too customer obsessed.” That’s by design.
So how does Amazon not become a charity?
Two guardrails. Not counterweights. Guardrails.
Guardrail #1: Frugality
“Accomplish more with less. Constraints breed resourcefulness, self-sufficiency, and invention. There are no extra points for growing headcount, budget size, or fixed expense.”
Frugality doesn’t limit how much you give the customer. It constrains how expensively you do it.
This is a critical distinction. Amazon doesn’t say “stop giving value to the customer.” Amazon says “find the cheapest, scrappiest, most inventive way to give value to the customer.”
- AWS didn’t build fancy custom data centers to be generous. They used commodity hardware and passed the savings to customers
- Prime’s free shipping wasn’t “let’s eat the cost.” Amazon renegotiated carrier contracts, built their own logistics network, and drove volume economics until the math worked
- Kindle hardware at cost wasn’t charity. It was a calculated bet that cheap devices would drive ebook purchases
Frugality is what makes Customer Obsession sustainable. You obsess over the customer, but you do it with discipline and resourcefulness.
Guardrail #2: Ownership (Long-Term Thinking)
“Leaders are owners. They think long term and don’t sacrifice long-term value for short-term results.”
This is the guardrail that makes the “cannibalization” plays rational business strategy instead of charity.
Every example I gave you in the table above (Marketplace, AWS price cuts, Kindle, Prime) all look insane in the short term. You’re giving away margin. You’re inviting competitors onto your platform. You’re selling hardware at cost.
But through the lens of long-term thinking, every single one of those decisions created enormous business value. Not in the next quarter. Not even in the next year. But over 5, 10, 15 years.
This is the Amazon Flywheel in action:
Lower prices → More customers → More sellers → More selection → Better experience → More customers → More volume → Lower costs → Lower prices → …
Customer Obsession is the business strategy. Frugality makes it efficient. Ownership makes it long-term rational.
The “charity” concern disappears when you understand this. Amazon isn’t giving things away out of kindness. They’re making long-term bets that obsessing over the customer will create more value than extracting maximum margin in the short term.
And they’ve been right. Consistently. For 30 years.
Why This Matters in Your Interview
Here’s why I’m telling you all of this.
If an interviewer ever asks you (and they will): “How do you balance customer needs with business needs?”
Most candidates say some version of: “Well, you have to find the right balance between what the customer wants and what the business needs.”
That answer sounds reasonable. It sounds mature. It sounds balanced.
At Amazon, it’s the wrong answer.
The Amazonian answer is:
“They’re not in tension. Obsessing over the customer IS the business strategy. The question isn’t whether to prioritize the customer. It’s how to serve them in a way that’s frugal and builds long-term value. Every time Amazon has bet on the customer over short-term margins (Marketplace, Prime, AWS pricing) it’s created enormous long-term business value.”
That answer will make a Bar Raiser’s ears perk up. Because it shows you don’t just understand the words “Customer Obsession.” You understand the philosophy behind it. You understand why it’s LP #1. You understand the flywheel.
The Customer Obsession Story Framework
Alright, let me get practical. If you’re preparing for an Amazon interview and you want to nail Customer Obsession, here’s the framework I recommend for structuring your stories:
| Step | What to Include | Dimension |
|---|---|---|
| 1. Customer Signal | A complaint, a data pattern, a behavior you observed | Working Backwards |
| 2. Dig Deeper | You investigated beyond the surface. The real problem was different from the stated problem | Anticipate the Unarticulated |
| 3. The Hard Call | You advocated for the customer against internal resistance: a delayed roadmap, a sacrificed metric, a difficult conversation with leadership | Cannibalize Yourself |
| 4. Frugal Execution | You found a resourceful, cost-effective way to solve it | Guardrail: Frugality |
| 5. Long-Term Impact | Measurable customer outcome AND long-term business value | Guardrail: Ownership |
Let me show you what a complete story looks like using this framework:
Complete example:
“We were getting support tickets about our API response times being slow. That was the customer signal. [Step 1: Customer Signal]
But instead of just optimizing the API, I pulled six months of usage data and discovered something interesting: 80% of our API calls were customers polling for status updates. They weren’t slow. They were frequent. The real problem was that we didn’t have a webhook system, so customers had to keep checking manually. They’d built polling loops that hammered our API. The ‘slowness’ was a symptom, not the disease. [Step 2: Dig Deeper]
I proposed building a webhook notification system. My manager pushed back because we had a feature release scheduled, and this would delay it by two sprints. I made the case that fixing the root cause would reduce support tickets, reduce API load, and fundamentally improve the developer experience. I brought the data. Leadership approved it. [Step 3: The Hard Call]
Instead of building a custom solution, we used an open-source event system that we adapted to our needs. It took us one sprint instead of three. [Step 4: Frugal Execution]
Support tickets dropped 70%. API load dropped 50%. And the feature we delayed? We shipped it two sprints later and customers barely noticed the delay, but they immediately noticed the webhooks. Our developer NPS for that integration went from 32 to 67. [Step 5: Long-Term Impact]“
That’s a story that hits all three dimensions and both guardrails. That’s a story that gets you hired.
Common Mistakes Candidates Make with Customer Obsession
Before I wrap up, let me flag the patterns I see candidates fall into. These are the traps. Avoid them.
Mistake #1: Confusing Customer Service with Customer Obsession
Customer service: Someone reports a bug, you fix it. That’s reactive. That’s expected. That’s every company.
Customer Obsession: You discover a pain point before anyone reports it, you fight internal resistance to prioritize it, and you measure the impact on the customer’s life.
If your story doesn’t have a moment where you went beyond what was expected, it’s not a Customer Obsession story.
Mistake #2: Confusing Customer Obsession with Earn Trust
I see this constantly. Candidates tell a story about building a great relationship with a stakeholder. They call it Customer Obsession. Interviewers hear it as Earn Trust.
The distinction is clear:
- Earn Trust: I built credibility and rapport with people
- Customer Obsession: I discovered a customer need, worked backwards from it, and delivered an outcome that measurably improved their experience
One is about relationships. The other is about outcomes.
Mistake #3: Forgetting That Internal Customers Count
If you’re a TPM, an infrastructure engineer, a platform team member, you might panic because you don’t have external customer stories. Don’t.
Internal customers absolutely count. But you must show the chain to the end user.
Weak: “My customer was the engineering team, and I helped them deploy faster.”
Strong: “The engineering team was my direct customer. By reducing their deployment time from 4 hours to 20 minutes, they could ship customer-facing fixes the same day instead of next week. End-user reported issues dropped 30% within the first month.”
Always connect the dots to the end user. That’s what makes it Customer Obsession and not just “I helped a coworker.”
Mistake #4: Missing the “Sacrifice” Signal
The strongest Customer Obsession stories include a moment where you sacrificed something for the customer. Your team’s roadmap. Your sprint velocity. Your quarterly metrics. Your own convenience.
If everything in your story was easy and everyone agreed, it’s not demonstrating obsession. Obsession means you cared enough to fight for it even when it was costly.
Mistake #5: Ignoring the Competitor Dimension
The full Leadership Principle text says: “Leaders start with the customer and work backwards. They work vigorously to earn and keep customer trust. Although leaders pay attention to competitors, they obsess over customers.”
Most candidates ignore that competitor line. But showing competitive awareness in service of the customer is a differentiator.
Example: “I benchmarked our onboarding experience against our top three competitors and found we were requiring 8 steps where they required 3. Our customers were getting a worse experience, and I used that competitive analysis to make the case for a redesign.”
But here’s the important nuance: there’s a difference between learning from competitors and obsessing over competitors. Competitors are useful as a benchmark for improving what already exists. But if you let competitors drive your roadmap, you’ll always be playing catch-up. You’ll build faster checkout lanes because that’s what everyone else is doing. You’ll never invent Just Walk Out.
The LP text is precise about this. It says: “Although leaders pay attention to competitors, they obsess over customers.” That word “although” is doing heavy lifting. It means: yes, be aware. But don’t let competitors be your compass. Your compass is the customer. The best innovations at Amazon never came from watching what Google or Walmart were doing. They came from watching what customers were struggling with.
The Complete Picture
Let me bring it all together.
Customer Obsession at Amazon isn’t a single behavior. It’s three dimensions operating simultaneously:
- Work Backwards: Start from the customer need, not from what you can build
- Anticipate the Unarticulated: Solve pain the customer has learned to tolerate
- Cannibalize Yourself: Make the hard call when customer needs conflict with your convenience
And it’s kept sustainable by two guardrails:
- Frugality: Obsess over the customer cheaply, resourcefully, inventively
- Ownership: Think long-term. Customer obsession IS the business strategy
Customer Obsession isn’t balanced by another principle that says “don’t be too customer-focused.” It’s #1. It’s deliberately unbalanced. And that’s exactly what makes Amazon, Amazon.
A Personal Note
I’ll be honest with you. When I first joined Amazon, I thought I understood Customer Obsession. I’d been customer-focused my entire career. I cared about users. I listened to feedback. I built things people wanted.
But it took me a while to understand the depth of it. To understand that it’s not just about listening to customers. It’s about thinking ahead of them. It’s not just about solving their problems. It’s about being willing to sacrifice your own comfort to do it. And it’s not in tension with the business. It is the business.
That shift in thinking changed how I approach everything. Not just interviews. But how I build products, how I lead teams, how I make decisions.
And I think that’s what Amazon is really looking for when they assess Customer Obsession. They’re not looking for someone who’s “nice to customers.” They’re looking for someone who has internalized this way of thinking so deeply that it’s how they naturally operate.
If you can show that in an interview, through stories that demonstrate all three dimensions, you won’t just get an offer. You’ll walk in there knowing you actually embody the principle.
And that’s a much better feeling than just passing an interview.