How to Know Your Customers by Amazon Senior Product Manager

The defacto for every Product Manager is to know his/her customers before building a successful product. While this may seem a bit too generic, there are finer details and techniques involved in analyzing the vast sea of customer reviews including the significant changes to a product, to complete the vicious cycle of Product Development.

In this session, Will Najar educates us on the strategy of “working backwards” and provides a deep insight into all segments of building the right product.

Customer Focused – Will Najar

profile photo Will Najar

Will Najar is a great orator apart from being a full-time Senior Product Manager – Technical, at Amazon. He has profound knowledge of being a successful Product Manager, who handles his customers with utmost patience. At Amazon, he is in charge of building systems, strategies, and policies that contribute to making reviews on Amazon trustworthy.

Knowing Your Customers

In order to provide an insight into the need for Customer satisfaction and product management, Will begins his talk by citing a winsome quote:

“One thing I love about customers is that they are divinely discontent. Their expectations are never static – they go up. It’s human nature. We didn’t ascend from our hunter-gatherer days by being satisfied. People have a voracious appetite for a better way, and yesterday’s ‘wow’ quickly becomes today’s ‘ordinary’ … You cannot rest on your laurels in this world. Customers won’t have it.”

Angry cat

He then mentions that it is important to know the right techniques to build a successful product. One of them is…

Working Backwards

Working backwards means starting from the customer, and interacting and collecting information from them regarding their needs and demands. This is a significant step and the research needs to be extensive and precise at the same time.

Next, by filtering out the redundant data, we refine the core issue. This leaves us with a roadmap to the end product development. Accordingly, we ought to plan the MVP and build the product. 

The cycle of delivering the most viable product is a persistent one. The product always gets better with customer feedback. There is no “flawless” one. There is always room for betterment.

This method of “Working Backward” reduces the disparity of the product with the market demands & hence reduces the chances of the product failure.

Will suggests that we consider the following points to substantiate the research.

  1. Who is your customer? 
  2. What is the customer’s problem? or What’s the opportunity you can hit on?
  3. Why does it really matter to the customers?

A Case Study: Uber

Toy car

In order to stress the importance of extensive research on the customer, Will takes the example of Uber. 

As a part of R&D on customer requirements to design policies and strategies, Uber came up with a few customer segments in 2008. However, as a responsible cab service enterprise, it should have considered on the segment for drivers’ welfare policies & development. But that wasn’t the case. 

The other competitors, however, did provide better Driver policies than Uber. The sudden drop in the Uber drivers & rise in the peer competitors hit the business graph gradually. This pushed Uber to revamp their information and include the “Uber Driver” as an important customer segment. Hence, before diving into the product finalization, craft out the customer segments. Consider all in the circle, to see no loss.

To put it in simple words, a business has two great features, the first being to deeply understand the customer need.

This is a critical step in product development. Understanding the customer through interaction is not sufficient. Empathizing with the customer for their needs plays an important role.

In this regard, feedback plays a major role. As Will says, there is a large sample set in our day to day lives, where we can expect valuable feedback on the products we develop. He quotes an example of his flight journey, where a co-passenger was criticizing Amazon as a product. He says, the criticism indeed helped him to invest back in development.

Some of the common means of feedback are:

  • Direct feedback
  • Feedback forms 
  • Customer reviews 
  • Sales conversations 
  • Social media posts 
  • Customer service 
  • User research 
  • Complaints 
  • Thank you notes 
  • Press coverage 
  • Product forums 
  • Emails 

Considering the above ways of obtaining the customer reviews, the volume of data that is expected to arrive is quite overwhelming. Not all reviews are worth our time. In order to simplify this process, Will encourages us to perceive the user feedbacks as anecdotes/stories. From the soup of feedback, you need to find the customer anecdotes that make sense, that bring value.

Ways to Analyze Reviews

Divide these anecdotes into themes. All reviews that belong to a theme must be sorted into a group. Ensure that each theme is workable. That is, it should help you identify the problem and craft a solution. 

One of the tools that help you narrow the reviews to the problems is the “Five Whys!”. Where you keep asking the customer Why? until you reach the core issue. This helps you to avoid the assumptions/misconceptions.

People looking at laptop

Assumptions Are Detrimental

While analyzing the feedback, more often than not, we tend to assume the problems/solutions for the customers. Will explains this with the Twitter case study, where an assumption led to an irrelevant development. 

A Case Study: Twitter

Analyzing customer data to produce a better product, has been the storyline of Twitter. After working with the statistics and math, Twitter came to an understanding that many customers refrained from tweeting, due to the limited characters of 140.

Following thorough groundwork, it increased the word limit to 280. However, this was a controversial development, as many fanatic customers claimed the breach of its ethos!

Customers are tricky!

This brings us to our second great business feature; to find an elegant way to meet the need.

Understanding the customers alone will not complete the business cycle. It is just the first step. It is rather crucial to implement the changes in an elegant way so that the customers accept it positively. In this process, it is important to embody certain principles and humility, is one of them.

Practice Humility

Though, this is not a commonly seen trait of many people, it is a crucial emotion for a PM.

Very often, the Product managers fail to accept the fact that they are not always the “typical” customers, even in case they are regular customers. Will claims this trait as “false consensus effect”, which often misleads development.

The humility to accept this fact and thus reduce their personal influence on the product can prove to be a significant contribution to product viability. The humility required can be attained through some of the following tactics:

  1. Leave your ego at home.
  2. Empathy
  3. A/B Testing
  4. Building a Minimum Viable Product


Once the product is built with all the contributing variables, Will imposes the need for extensive product testing under various environments. This often reduces the probable faults that might occur at the customer end. The factors for testing are mentioned below:

  • Experimentation 
  • A/B or multivariate testing 
  • Product metrics 
  • Beta testers 
  • Soft launches 
  • Phased rollout 
  • Minimum viable product (MVP) 
  • Prototyping 
  • Launch fast, fail fast 
Man looking at mindmap

Key Takeaways:

  • Understanding the customer plays a major role for successful product development.
  • The two emotions a Product Manager has to embody, to know his customers better are:
    • Empathy
    • Humility
  • Working backwards from Customers -> Product proves to be fault-proof way for Product Development.

If you liked this talk, you can find many more on our YouTube channel, and you can also join our global Slack channel for Product Managers everywhere.

ProductCon NYC banner

Enjoyed the article? You may like this too: