Amazon Uses These 3 Strategies To Make You Spend More

A few weeks ago, I was shopping for a new phone case.

I found a nice one on Amazon that was simple, durable, and just what I needed.

I was scrolling down to read reviews when I saw something right below it.

Customers who bought this item also bought

It showed a screen protector. I paused. My old screen protector had cracks.

Without even thinking, I added both to my cart.

What started as a ₤10 phone case turned into a ₤18 order.

And here’s the thing. I didn’t feel tricked. I felt helped.

It was like Amazon read my mind and knew exactly what I needed before I did.

Sound familiar? We have all been there, looking for one thing but then buying multiple products, and it all makes sense.

But how does Amazon do this so well?

Let’s understand.

The Idea Behind “People Also Bought”

Amazon’s People Also Bought is a product recommendation system. Instead of showing random suggestions, it analyses customer buying data to predict what you might need later.

For example:

• Amazon might suggest a wireless mouse and laptop sleeve when you buy a laptop.
• If you buy a coffee machine, it might recommend coffee pods and a milk frother.

When this feature works, it not only helps customers find more relevant products. It also helps Amazon’s business by:

• Increasing the average order value (you spend more per purchase).
• Improving the customer experience (you get what you need).
• Building trust (you feel like you are making a wise decision).

The beauty is that it’s not pushing you to buy more (like a clingy salesman). It’s only guiding you toward the right choices.

Now, let’s look at how Amazon made this work at scale.

How Amazon Built This Billion-Dollar Feature?

Amazon didn’t invent product recommendations, but it perfected them.

The Customers Also Bought feature is a carefully engineered system that relies on data, algorithms, and behavioral psychology to encourage customers to buy more without feeling pressured. Here’s how this recommendation engine works:

1. Collaborative Filtering: The Secret Behind the Recommendations

Amazon’s recommendation system starts with data.

It records every action you take on the platform, from browsing to adding something to a cart and completing a purchase. But it doesn’t just look at individual purchases.

It analyzes millions of shopping patterns.

• Tracks millions of purchases: Amazon’s system continuously monitors what products people frequently buy together. If thousands buy a laptop and a specific wireless mouse together, that pairing becomes a strong recommendation.

For example, when I am trying to buy a book, the frequently bought section shows this because people who bought the book I wanted to buy bought these together:

• Groups similar shoppers: The algorithm groups people based on shopping behavior. If customers like you tend to buy certain combinations, those products are likely to be recommended to you.

For instance, when I searched for novels, Amazon immediately pushed me into a group of people who did a similar search to mine.

• Adjusts based on data: If a new product starts trending alongside a popular item, it quickly gets included in recommendations. If an old product stops selling well, the recommendation system removes it.

This is called collaborative filtering, and it’s the backbone of Amazon’s success.

It works because most brands recommend products based on similarities (e.g., If you like this book, you might want another book). But Amazon goes deeper.

It looks at what customers bought together, even if the products are unrelated.

That’s why you don’t only see more books when you buy a novel. You might also see a cozy reading lamp, a notebook, or a coffee mug.

2. Contextual Timing: Amazon Knows WHEN You Are Most Likely To Buy

Timing is everything. Amazon doesn’t just show recommendations anywhere. It carefully chooses the perfect moment when you would say YES.

So, it strategically places People Also Bought for maximum impact:

• While browsing: Customers Who Viewed This Also Viewed shows up to spark curiosity.
• Before checkout: When you add an item to your cart, People Also Bought appears. It makes you think, Wait, do I need this too?
• At checkout: Frequently Bought Together appears, nudging you to upgrade your purchase at the last second.

Say you are buying a set of kitchen knives. You add them to your cart. Just before checkout, Amazon might suggest a cutting board, knife sharpener, and recipe book.

For example, I added a kitchen knife set to the basket and these were the recommendations before I checked out.

At that moment, adding an extra item when you are already committed to buying something feels like a no-brainer decision, not a big one.

That’s why Amazon shows recommendations when you think you might need them. They designed each step strategically to remove friction and increase spending.

3. Continuous A/B Testing & Machine Learning

Amazon never settles on one version of its recommendations. They change in real time based on what’s working. Here’s what makes their system incredibly effective:

• Real-time A/B testing: Amazon constantly tests various combinations to see which gets more sales. If one product pairing performs better, it gets prioritized.
• Seasonal adjustments: Recommendations shift based on the time of year. In December, holiday gifts dominate suggestions. In summer, outdoor gear takes over.
• Personalized learning: If you have bought fitness gear in the past, Amazon is more likely to suggest customized workout gear.

If Amazon notices that people buying wireless earbuds are suddenly also buying carrying cases more often, that product pairing gets pushed up the list so Amazon never misses a trend. And all of this happens automatically through machine learning.

Amazon keeps learning and fine-tuning its system so customers see only the best-performing suggestions.

Lessons for Product Managers

Amazon’s recommendation engine is about making the shopping experience feel easy and intuitive for the customer. And to make more money for the business.

It is a perfect example of how user needs can be balanced with business goals (without frustrating the users .)

Here’s what we can learn from it:

#1. Timing Matters - Offer Solutions When Users Need Them Most

Amazon succeeds by suggesting products exactly when users are ready to buy.

Do this: Understand when your users most need help or information. Provide support, features, or reminders at that exact moment to make their experience better.

#2. Look at How Users Behave, Not Just What They Say

Amazon’s strength comes from tracking what customers actually do, not what they claim they'll do.

Do this: Observe how users interact with your product in real-life situations. Improve your product based on actions you see users taking, rather than relying only on surveys or guesses.

#3. Regularly Test Small Changes and Improve Quickly

Amazon continuously tests and updates its suggestions based on what actually works best for users.

Do this: Frequently test small improvements with users. If something works, adopt it quickly. If something doesn't help users, stop doing it right away.

#4. Users Trust Products that Feel Helpful, Not Pushy

Amazon’s recommendations work because they feel helpful, not forced.

Do this: Always communicate clearly how your product helps users. Avoid making users feel like they're being sold to. Instead, focus on solving their problems or making tasks easier.

Final Thoughts

Amazon’s Customers Also Bought works because it feels like a natural part of shopping.

It doesn’t feel like a dirty sales tactic. Instead, it makes customers feel like they are making better, smarter buying decisions. And that’s the real magic.

If you get your recommendations right, users won’t feel like they are being sold to. They will feel like you are helping them make better decisions.

As a result, they don’t just buy more, they trust you more.

Have you ever bought something based on Amazon’s recommendations? If so, what made you do it?