Peter Yang’s 10 rules for making products that customers love

Whatever your social media platforms are, there’s a fair chance that Peter Yang has built a product for them. His career stretches back over a decade, starting with an early career pivot from product marketing into product management during a three-year stint at Meta. He went on to build products at platforms such as Reddit, Twitch, and X (formerly Twitter). In the past few years, Peter has shared many of the opinions, insights, and experiences through a popular newsletter read by tens of thousands of tech professionals. He’s also built a wide-reaching following on social media, where he’s traded tweets with Elon Musk, roasted AI, and taught product lessons through cat memes.

Here, Peter shares 10 lessons on the craft of product management, starting with personal growth to influencing entire teams or organizations, all in service of building products that customers love.

1. Good “product sense” isn't intuitive, it needs to be cultivated

In the world of product management, product sense is one of the most valuable skills. The problem with this idea is not only is it vague, it also implies that a PM can just build their product sense once and keep it forever. I just think that's wrong. The market is changing all the time. Customer needs are changing all the time. You need to have the humility to continue to understand and improve your product sense through cultivating empathy, creativity, and craft.

2. Use empathy to diagnose the problem

If you’re going to make something customers love, you need to get really clear about what they need. A lot of people skip this critical step and rush into coming up with a vision or solutions. You must resist this instinct. As Tony Fadell, who served as Senior Vice President of Apple’s iPod division and co-founder of Nest Labs, says, “A PM’s superpower is empathy.” You build empathy by treating your customers like team members. You can also spend a day—or even an hour—using your product, just like a customer would.

3. Before you get lost in the day to day, pave the way with big ideas

After you’ve diagnosed the customer and business problems, don’t jump into day-to-day execution right away. Instead, think big by tapping into your team’s creativity to set the mission, vision, and strategy. Huddle with your team to discuss the problems and brainstorm blue-sky ideas that you can then prioritize. Figure out a simple solution—say, three steps—that really meets the customer’s needs the best. As you learn more about the customer, you will need to make trade-offs, and potentially shift priorities.

To shape your vision, try this vision and strategy artifacts template from Coda.

4. Know when to hit the brakes, change direction—or even pull the plug

Quality can sometimes require making difficult trade-offs. For example, choosing quality might mean missing a launch date or a goal metric in order to ship a delightful experience.

At a past company, we had an OKR to improve a metric by releasing a feature in a quarter. However, as we talked to customers, I realized adding an additional feature would make their experience a lot better. Since we didn’t have enough time, deciding to build it meant that we had to miss the deadline. OKRs are intended to help move the business forward. Sometimes, stakeholders will understand this means missing a short-term milestone for greater success in the future. In my example, that extra feature was the thing that a lot of the customers pointed out made all the difference in their experience. Craft is really about going that extra mile, and being obsessed (in a good way) with all the different details, trade-offs, and bug bashing that helps you deliver something really worthwhile.

Sometimes, what’s best for the business is going in a new direction altogether. When I was at Reddit, I was working on Reddit Talk, a live audio product that millions of people used and loved. They would jump on, bond with strangers, laugh together, cry together. The economy took a downturn, however, and we had to cancel the project because it didn’t grow the company metrics that mattered. It requires a certain level of maturity to step outside the bubble of just trying to grow and improve a product and think about the bigger picture, too: “Is this the most important thing to work on right now for the company?” From the perspective of Coda Co-founder and CEO Shishir Mehrotra’s PSHE framework, I did a good job of defining the solution, executing on it, and even understanding the customer problem. People want to get together, live. But I didn't do a good job of thinking about the business problem.

5. Automate the work that gets in the way of work with AI

Your job as a PM is not to craft the perfect OKR, internal doc, or product review deck. Those are intermediate steps. Your job is to deliver value to customers and grow your business. If you find yourself spending just as much time on these intermediate artifacts as building the actual product, it’s time to take a step back and adjust. Consider delegating the tasks, or automating them with AI.

Using AI is very natural for PMs. If you think about it, what a PM does every day is like prompt engineering. It’s communicating in a way that empowers other people or teams.

AI is not so good at coming up with original work, but it’s great at synthesizing and summarizing. I like using FigJam AI to organize and remove jargon from brainstorms. I also use AI to summarize many pieces of customer feedback, and to make product requirement docs or strategy docs clearer and more concise.

Spin up a product requirement document quickly with this FigJam template.

6. The talent stack is collapsing, so learn to wear multiple hats

With AI making builders more productive, as well as the difficult economic times in the tech industry, the job market now favors builders over managers. Product managers need to be able to prototype and bring products to life. To adjust, learn how to work across functions and collapse the stack. Be a PM who can write copy, a designer who can code, and a developer who can drive growth. Personally, as a non-technical PM, I’ve been trying to build an app with AI.

Because smaller teams can move a lot faster than huge organizations, I hope to see more companies elevate the individual contributor product manager, someone who can get a lot of stuff done and deliver great products without having to build a team or manage anyone. That totally exists on the engineering side; there are technical architects and directors who have no reports. Staff- and principal-level PM roles focused on managing products, not huge teams of people, are definitely gaining greater importance. Small teams are in, and large teams are out.

7. Be aware of the shadows of your superpowers

Meta’s Vice President of Product Nikhyl Singhal says, “Every superpower also has a weakness that needs to be addressed to advance your career.” For example, a great storyteller might struggle to get into details, while an entrepreneur might have trouble delegating to others for execution. These weaknesses are your shadows. If you leave them unchecked, they can hold back your career. For example, I care about crafting quality products and getting stuff done. But sometimes, I can move too fast instead of bringing people along.

It’s challenging to permanently fix a shadow, which is more like a personality trait. Without it, you don’t have your superpower anymore, either. To address it, you must have the desire to change, and the humility to ask for—and listen to—feedback. You can approach your shadow by being very public about it. For example, you can have a strong point of view while making it clear to your team that you’re open to changing your mind: “I’m not always right. You should disagree with me.” If you’re transparent about your shadows, this understanding empowers the other person you work with to tell it to you straight, as opposed to leaving some feedback behind the scenes or telling your boss.

8. Make customers part of your team

The best way to build empathy for your customers is to make them an extension of your team. Recruit a few dozen customers who are a mix of beginner, intermediate, and advanced users to be in conversation. These people should be early adopters of your products and features who will give you passionate feedback and eventually become advocates who help spread the news to other customers.

Once you’ve recruited your customers, create a dialogue with this community-led product development framework:

Create a community. Announce your product early and invite customers to join an online community (e.g., Slack or Discord). Ask them to introduce themselves and start building trust.

Build in public. Ask customers about their pain points and share product ideas and designs early and often. Give them product demos and show them how you’re making adjustments to your roadmap based on their feedback.

Build a casual environment. Make it easy for customers to talk about anything. This is the best way to uncover new pain points and walk in your customer’s shoes.

9. Make great decisions async with a decision doc

One of the best ways to protect your craft time is to learn to make great decisions asynchronously. If you learn how to do this well, you’ll save your team from endless hours of meetings, and they’ll love you for it. After all, nobody dreamed of sitting in Zoom meetings all day as a kid. Here’s my process:

Identify key people. Ideally one decision maker and a couple of stakeholders.

Make one decision doc for communication. Keep all the discussion in that doc or a corresponding Slack group channel, and tag relevant people.

Make great decisions async with this decision doc template.

Set context. In less than two pages, describe the decision, your recommended option, and any trade-offs with other options.

Use numbered lists for discussion. People can refer to numbers for points, like, “I like option three because of XYZ.”

Push for a decision and share it broadly. Write something like, “It sounds like people prefer number one, any strong objections to moving forward?”

Meet when you need. When a decision feels ambiguous, is hard to reverse, or the discussion isn’t moving forward, you need a meeting.

10. Remember why you got into this work in the first place

A lot of my writing and interviews focus on why many PMs got into tech, which is usually to craft really awesome experiences that exceed customer expectations. It’s easy to lose that focus, though. So many PMs have to contend with competing demands, like stakeholder expectations and OKRs, that can sometimes overshadow the joy of actually building something meaningful. Not everyone is a fit to manage a 200-person product org, or to try to make it as high up the ladder as they can. Like many PMs, I joined tech to craft products that customers love. I never want to lose sight of that.