I recently picked up the book Working Backwards and honestly I loved it. It gives a peek inside how Amazon works, not in a dramatic or flashy way, but in a very practical, real sense. The book talks about the systems, the thinking, the decision making. A lot of it made me think about how we build products and run teams.
So I wrote down some notes and thoughts that stayed with me. Sharing them here in case someone else finds them useful.
1. Start with the Customer and Work Backwards
This is a big one. It’s actually where the book and the blog gets its name. The idea is that you don’t start with a business idea or a technical solution, you start by deeply understanding the customer’s problem.
What do they need? What’s frustrating them? What would make their life easier?
Then, once you have figured that out, you work backwards to build a solution. You think about what kind of product would solve that, how to make it happen within your own limitations (team, time, money, tech) and then go from there.
It’s a simple concept but very easy to forget. A lot of teams (and I have done this too) just build things because they can, not because someone asked for it or needed it.
2. Obsess Over the Customer
At Amazon, they are not just “customer-focused”, they are obsessed. It’s a mindset where every decision is rooted in how it will affect the customer.
And it’s not always about asking customers what they want. Sometimes customers don’t even know. It’s about digging deeper.
What are their pain points? What do they complain about quietly? What slows them down?
For example: if you are building an online form and 20% of people drop off halfway, you don’t just say “that’s normal”, you ask why. Is it too long? Confusing? Annoying? That’s where customer obsession kicks in. You fix the little things that people didn’t even know could be better.
3. The bar raiser thing
One hiring practice they talk about is this idea of “raising the bar.” Basically, every person you hire should be better than the average current team member in at least one meaningful way.
Not better at everything. Not perfect. But they should bring something new, something they are great at. Maybe it’s design, maybe it’s speed, maybe it’s empathy. It keeps the team growing and evolving.
I really liked this. Not because it’s about hiring “only the best” (whatever that means) but because it’s about thinking hard about why someone should join your team and how they make it better.
4. Single-threaded ownership
This one made a lot of sense to me. The idea is simple: if something’s important, give it a single owner. One person whose full time job is to make sure it succeeds.
Too often, we spread things across 3 different teams, or assign it to someone already juggling 5 other things. Then we wonder why it moves slowly or gets stuck.
Giving one person full ownership and the freedom to focus, just works. It brings clarity and accountability. If something matters, it should have a name next to it.
5. PR/FAQ method
One of their most famous practices is writing a fake press release and FAQ before building a product.
The press release explains what the product does, why it matters, and how it helps the customer as if it’s already launched. The FAQ then answers common questions and challenges, like how it works, what might go wrong, or what edge cases need to be handled.
This is such a smart exercise. It forces you to think from the customer’s point of view and make sure the product is even worth building before you get too deep.
It’s also a great way to align the team. Everyone reads the same doc and understands the vision upfront.
This part hit home. Most of us track output metrics things like revenue, user growth, engagement. But Amazon puts a lot more focus on input metrics, the things you can control that lead to those outcomes.
For example: if your goal is more signups, don’t just look at the signup number. Look at things like page load time, how clear your CTA is, whether people drop off at step 2. These are inputs you can act on.
Fixing those leads to better outputs naturally. It’s more actionable and more motivating.
7. The flywheel idea
There is this flywheel diagram Jeff Bezos drew back in the day. It shows how improving one part of the business feeds into others.
A better customer experience brings more traffic
More traffic attracts more sellers
More sellers means better selection
Better selection improves experience
And around it goes
It’s this loop that feeds itself. The key is identifying what your version of that flywheel is and figuring out which part you can push to make the whole thing spin faster.
The book ends on a very real note. You don’t need to do everything Amazon does. In fact, you shouldn’t.
But you can take the ideas that make sense for you. Try them out. Adapt them to your team, your culture, your goals. Whether that’s writing PR/FAQs, focusing more on input metrics or just asking better questions.
If you have read Working Backwards too, would love to hear what you took away from it. And if you haven’t read it yet, it’s definitely worth picking up.
