Info iconLetter I in a circle

By continuing to use this site you consent to the use of cookies in accordance with our cookie policy

Edmond Lau and Jean Hsu on engineering more leaders and better working relationships

Senior Director of Engineering, Intercom

Louis Bennett

@louisbennett

As up and coming engineers in Silicon Valley, Jean Hsu and Edmond Lau long sought more influence and impact in their respective positions – but they couldn’t shake the feeling of being stuck.

Their story isn’t uncommon. Many engineers feel like they must make a jump into management to lead. But as Jean and Edmond will now tell you, following years of engineering leadership at Quip, Quora, Medium and more, that simply isn’t the case.

They’ve now started a company together, Co Leadership, to help pave the road for engineers to become leaders. After coaching 100+ tech leads, managers, directors, VPs, and CTOs, Jean and Edmond are distilling their most valuable lessons into workshops and online courses.

I hosted Jean and Edmond on our podcast to get their take on the difference between leadership and management, the power of positive feedback, and why trust is the foundation of the most effective engineering teams. If you want more insights from Edmond and Jean, they’ve actually put together a free guide for Inside Intercom listeners, which is all about designing better working relationships.

What follows is a lightly edited transcript of our conversation. Short on time? Here are five quick takeaways.

  1. Leadership is a mindset, not a title. It’s knowing the impact and the influence that you want, and feeling empowered to take steps toward achieving it.
  2. The foundation of a positive, two-way working relationship is having a sense for what motivates your teammate. Know what drives them in turn drives trust and shared success.
  3. If you notice a gap in your organization or something that could be improved, speak up. It’s likely that you’re not alone in your opinion or experience – a realization that can spark change.
  4. Too often, post mortems and retrospectives focus on what went wrong with a project, rather than what went well. If people don’t know the impact of their positive actions, they might stop doing them.
  5. Engineers inherently know that when tech debt piles up, you need to invest in fixing it. The exact same is true for relational debt.

Louis Bennett: Jean and Edmond, welcome to Inside Intercom. Can you give us a feel for what led you both to Co Leadership and explain your mission there?

Jean Hsu: We both spent about a decade in various engineering roles including IC, TechLead, and other leadership positions. Both of us had this experience where we felt stuck. We wanted to have more influence and impact than we were having, but we were trying all these things and we couldn’t figure out how to do it.

Once we both moved into engineering leaders, we saw this was a pretty common pattern. We really want to help people bridge that gap, make it an easier transition (into leadership) and help people live the lives and careers they dream about.

Edmond Lau: There were times earlier in my career where I wanted more leadership roles and I would do things like ask my manager to become a manager. I would be waiting for leadership and authority to be handed to me. When that didn’t happen, I just felt like, “What do I do now?”

It’s very easy for people to feel stuck and wait for opportunities to come. A lot of our work at Co Leadership is helping people feel empowered – to actually start leading from where they are.

Why anyone in your org can be leader

Louis: How do you define leadership?

Edmond: Leadership is having some sense of the impact and the influence that you want, and then feeling empowered to take steps toward moving towards that direction.

Jean: Leadership is really about your mindset. A lot of times people will wait for a leadership role to be granted to them or their potential to be recognized and then given this title. Then what they often hear is, “You need to display leadership before you can have the role.” They just don’t know what that means. I didn’t know what that meant and I tried a bunch of things without much success. Leadership is more about the mindset of taking initiative and being responsible for the changes you want to see.

Edmond: An implication of that mindset is that everyone really can be a leader. It’s not something that has to be attached to a title or a role. It’s something that you can decide and choose to take on. Then if you’re given the right tools, you can be more effective at doing that.

Leadership is not something that has to be attached to a title or a role

Jean: Especially with engineering teams. If you think of leadership just as management, there aren’t that many spots. As a result, you see a lot of people moving into roles like mid or senior level IC, a tech lead or tech lead manager. All these roles change very quickly. They’re not super well defined, so we really want to equip people with the skills to have leadership in these types of roles that aren’t as explicit as, “Your role is an engineering manager.”

Louis: There’s a certain kind of inherent tension to engineering management. Someone will have developed their skill set in one walk of life – computer science, for example – and then there’s so much organizational gravity pulling people towards a management role, which is a totally different discipline. But leadership is something you can show to be a senior individual contributor across all those things. It fits into that gap and can provide a strong foundation to switch from one to the other.

Edmond: There are a lot of different problems that exist in organizations where there are gaps that exist. Really anyone who builds awareness around that gap and takes initiative toward it can step up and be a leader, even without following those default patterns of what it means to be a manager or what it means to be a tech lead.

Lessons learned in coaching engineering leadership

Louis: Co Leadership offers training for people across a lot of different companies as well as deeper training sessions with specific companies as well. What do you think the most impactful things or the most surprising things have been that you personally have learned from the attendees of your courses?

Jean: For our independent events we have people come in pairs, such as an engineering manager and a product manager who have been working together for some time. We teach them some skills around how to discover what’s important to the other person. Then we send them off to have a five-minute conversation using these skills, asking open-ended questions and really listening. They’ll come back and say something like, “I just discovered more about what motivates this person. I didn’t even know that and we’ve working together for a year and a half.” That’s so rewarding for us to hear that we’re actually having that impact on people who work together all the time.

Edmond: It’s super motivating because if we can really spread the work that we’re doing at scale and teach a lot of these foundational, trust-building, relationship-building skills to more people, so many more teams would be able to achieve higher levels of success. If you’re able to really understand what’s important to all the stakeholders around you and people you work with, and you’re able to design much more fulfilling relationships at work, people would get so much more done and be way more efficient too.

Louis: It’s amazing how much of getting work done is just getting people to actually trust each other to work well together, to find areas where motivations are aligned or not aligned in order to help get stuff done versus just doubling down on, “We need to do this thing. Why aren’t we doing this thing?” It’s really about, “How do we actually work together to have shared accomplishments?”

It’s amazing how much of getting work done is just getting people to actually trust each other

Jean: That’s one of the reasons why this is a gap. As engineers we’re so focused on the how, the problem solving around a particular problem. It’s easy to focus on the tactical elements of something rather than taking a step back and investing in trust-building and discovering what’s important to people.

Louis: I’ve always approached this from the simple fact that I can’t read people’s mind. If something is important to someone, they’ll need to tell me that. I can infer, but the likelihood that I’ll get that wrong is also very high. If I say, “I think you’d be great at this,” they’ll need to tell me, “That sounds horrible. I don’t want to do that.” None of us will be right all the time in that way and so finding ways for people to effectively communicate, is empowering – whether it’s an EM and PM, peers or manager and direct reports.

Edmond: When we realize that we can’t read people’s mind it means both that we not only have to share what’s important to us, but we also have to ask people what’s important to them. It becomes this two-way, discovery conversation where you put what’s important on the table so you can actually start to make decisions around it.

An engineer is never alone in their problems

Louis: When you think about teams, beyond 1:1 conversations, have you found any frameworks that work well for teams to talk about things like this?

Edmond: That reminds me of the workshop we ran a while back with members of the engineering team from Medium. It was actually the reason why we started doing pairs at our workshops. There’s all this shared context from people who all work on the same team, but oftentimes what’s important to this group of people isn’t made explicit.

One of the workshop prompts was, “What are gaps that you notice in the organization or on the team?” We created this shared context of what gaps everyone saw and what they wanted their team to be. People were able to share a lot of things and build a confidence that, the gap they see on the team is something that everyone else sees too. They’re the one putting the words to it in this group, but there’s actually a lot of buy-in, alignment and support if they were to take initiative and tackle it.

Jean: Without that level of alignment it’s easy to think, “I’m the only person this is bothering. No one else is having this problem because no one’s bringing it up.” Then everyone’s sitting with this thing thinking, “I shouldn’t bring it up. I don’t want to be the one complaining.”

Edmond: That’s a broader limiting belief that we see among a lot of engineering teams – people think that they’re alone with this problem or that someone else will fix it. The impact of either of those beliefs is such that I feel less empowered to take leadership. Those end up being two of the beliefs that we work towards equipping people with the tools to overcome in these leadership experiences that we run.

The power of positive feedback

Louis: It feels like there’s some crossover with ideas like blameless post-mortems and things like that, of people being able to have shared vulnerability to say, “Here is exactly what happened. Let’s not talk about the who, but let’s talk about the circumstances that created this,” and really encouraging people to name the obvious to generate more insightful conversation that way.

One thing we do at Intercom that is very effective in promoting the greater good is helping people understand it’s not necessarily a people problem. Maybe it has to do with the circumstances of the problem or something we haven’t prioritized or some other extenuating circumstance. Those frameworks work best when people can build upon an existing strong relationships as well as a place to be safe.

When I think about your work, I think a lot about the research that’s been done about people performing better when they have psychological safety, and that vulnerability and psychological safety go hand-in-hand as well. Have you seen any direct or indirect experience that would confirm or refute that?

Jean: In one of the Medium workshops, we also had people go around and hear the impact that they’ve had on other people. We had everyone else in the group talk about the impact that person had on them. When we talk about blameless post-mortems, we usually think about when something’s gone wrong; a lot of my and Edmond’s work is focused on what’s going well, which does build psychological safety. Hearing 10 people say, “When you do this, it makes me feel inspired to do my work,” or when we do our post-mortems it’s mostly positive.

Edmond: We go through lessons learned, assessment of progress, and then gratitudes.

Jean: It’s focused on what’s going really well, and that builds up psychological safety. Then when we do have to give each other feedback in the moment we have this trust that we both really enjoy working with each other and are doing really meaningful work.

If people don’t know the impact they’re having with all their positive actions, they might stop

Louis: Do you find that certain personalities get that better than others? If I speak for myself, I’ve gone through performance reviews before where you hear, “Everything’s going well,” and I’m sitting on the edge of my chair waiting to hear about the things I could do better. I’m not saying I don’t listen, but that’s not the feedback I’m craving in some ways. Oftentimes people don’t provide enough or give enough credence to positive reinforcement feedback that way.

Edmond: It’s a very strong engineering bias. When we do code reviews we look for what’s wrong. It’s oftentimes very rare for people to say, “What’s going right?” in a piece of code, and we end up applying that same mentality to people. If 80% of the time things are going well, maybe 80% of the retrospective should be spent on celebrating what’s going right.

Jean: …Or figuring out how to do more of what’s going right, rather than focusing on the small percentage of things that are not going well.

Edmond: There’s a risk of not focusing on what’s going right: if people don’t know the impact they’re having with all their positive actions, they might stop doing it.

Jean: That’s an important point. When people share positive feedback, they often don’t talk about the impact of the positive things you’re doing. They’ll just say, “You did a good job,” or something that just seems vague and positive. They don’t say, “Because you did this, now so-and-so can really be unblocked,” or, “This person feels much better about their work.” A lot of times when you don’t have the impact shared, especially as engineers, we tend to just focus on, “What are the things I can change?”

Louis: If I think of every post-mortem I’ve done, if it’s divided into things that went well, things that went better, and things to change, the “things that go well” column is always given short shrift. It’s three or four different things and it’s nothing about the root cause of why things are going well. It’s more digging into, “Why didn’t that happen? What was the sequence?” From a healthy culture perspective you want to make sure you don’t end up messing up the good stuff while you’re trying to fix some of the things that could be going better too.

Edmond: There’s so much research in the field of positive psychology around how happiness leads to more success and better performance. Celebrating what’s going right can also just create a much happier team and also lead to better performance and success for the team too.

Building trust in geographically distributed teams

Louis: One thing that’s different at Intercom versus other companies I’ve worked for is the geographical distribution of the employee base. We’re in five different offices in four different countries, so being able to effectively work together across continents is predicated on trust and being able to say, “There’s an aspect of happiness within that.” Basically you, until you physically meet people in-person, are two dimensional in that space. You don’t necessarily get that character of, “This is who this this person is. This is how they feel about these things,” and the off-hand conversations that help build trust over time. I need to remember that if I have an urgent question for someone in part of the world that’s done at the office, I’m taking him or her away from their family. Just acknowledging that, or if they’re calling me at 5:00am, acknowledging that I might be waking up, makes it easier to work together too.

Jean: Remote or geographical distribution is a forcing function to make sure that you have explicit conversations that build trust. A lot of times teams rely or fall back on being able to see each other. You can get a sense of, “Is this person doing okay?” just by seeing them at their desk. The absence of that actually forces you to have these conversations more explicitly.

Edmond: One of the issues with remote teams is when the meetings are focused on making sure the logistics work as opposed to spending time getting to know one another – understanding what people’s core motivations are, what’s important to them and what makes them happy at their job.

There’s also a piece I think that would accelerate building trust: whenever issues do arise, clear up the stories that can get in the way. One of the researchers who really influenced our leadership experience work is Brene Brown. She does a lot of research into vulnerability, and how oftentimes the stories we make of people that we don’t know too well hold us back from stronger working relationships.

When there is conflict in remote working relationships, and there isn’t a way of clearing up these narratives that we make of other people, that’s when trust becomes harder to build. Spending time clearing up those stories helps build that trust a lot.

Jean: You can definitely get pretty imaginative with stories when you don’t see the person in the office. It’s like, what did that person mean when they said this on Slack? What did that email or code review comment really mean?

Louis: I think we’re all our own worst enemies when it comes to interpreting what other people said, either in the sense of giving people too much credit for what they’re saying and missing something nefarious. The much more common case is there’s some off-hand comment that we’re mentally chewing on for hours that doesn’t match the intention of what that comment was at all.

Jean: Meanwhile they were just trying to get through 20 pull requests and they were tired.

Reading and resources for engineering leaders

Louis: I’ve worked with a lot of engineers going through the transition into some variety of leadership role. There’s this sense of people waiting to dip their toes into it and there’s a chicken egg scenario of becoming a leader before you can become a leader. One of the things that I’ve done is recommend books that are helpful to me. What books and reading has been interesting to you to help people understand what engineering leadership and management means?

Jean: One of my favorites is Five Dysfunctions of a Team, which ties this conversation about trust. The author talks about these five dysfunctions that you really need to address in sequence. The basic one is lack of trust and building trust and how that can be the foundation for everything else. It’s written as a fable so it’s really easy to read.

Edmond: Another book that I really like that’s less focused on management is Extreme Ownership by Jocko Willink. He was a former Navy SEAL and ran the team in Iraq. A lot of that book is around this mindset he calls, “Extreme ownership.” Taking ownership of the outcomes.

Oftentimes we disempower ourselves when we believe it’s someone else’s responsibility

He talks about one training exercise where he broke his squad into a bunch of different sub-groups. They were running these races and they took the leader from the winning group and swapped them with the leader from the losing group, and the losing group was able to win. This person had embodied what it meant to take ownership over that group and was able to motivate everyone in that group to actually do well.

That book sticks with me because it has this mindset of, “You are in charge of the results that you want to see. It is not someone else’s responsibility.” Oftentimes we disempower ourselves when we believe it’s someone else’s responsibility.

Jean: It reminds me of another book that I found really influential, Turn the Ship Around. The author runs a nuclear submarine, and he changed the entire model from leader-follower model where there was a hierarchy of, “Do this,” and then that person would tell someone else what to do, to a leader-leader model. By the end he would be wandering around the submarine and people would come up to him and say, “This is the context. I’m going to make this decision. What do you think?” He’d go around approving things and everyone would come to him with the information he needed to make the decision. I thought, “If this guy can do it with a nuclear submarine, we can definitely do it with the internet and software.”

Louis: It puts it in context. Sometimes it feels like there’s no idea sharing in the way that software engineers do management, where we think about, “How do you break things into component parts and have distributed systems that talk to each other?” Management tends to be this talked down, orchestrated thing that has fewer of those components. That’s why I always like hearing about books that have ostensibly nothing to do with software engineering because it brings in new ideas.

Applying design to relationship building

Edmond: Translating some of these concepts that are well known in other industries into language that maybe engineers can better resonate with is something that makes us very excited about our work. How do we use our engineering careers and everything we’ve learned about abstractions or investments and infrastructure, and apply them to relationships?

Jean: Every engineer understands that when the technical debt gets bad you need to chip away at it. You need to invest in it. You need to create projects around it. The same is true for relational debt. If you invest in it upfront, then you can set yourself up for a better relationship down the road, but sometimes relational debt gets pretty bad and then you have to really invest a lot of time to make that working relationship better.

Edmond: We know that we definitely should design our software if we want it to achieve our desired goals, but we don’t apply that same mindset to our relationships. If we want our relationship to turn out a certain way, it would be helpful if we explicitly designed them.

Louis: Is there a point where it becomes awkward? Where you’re talking so much about your relationship and how you measure successes of relationships among teammates that it makes it a little impersonal in your experience?

Jean: Right now what we’re seeing an under-investment in talking about how people want to work together. I think we’re pretty far off from over-investing in there, and at least for us working together we definitely invest a lot of time in conversations just around what is and isn’t going well. It’s always surprising to me how well it’s paid off and made everything else easier.

Edmond: Any sense of awkwardness comes from using a new muscle that you just haven’t exercised before. It can be little uncomfortable when you don’t quite know how to do it correctly, and a lot of our day-long events are teaching people the language they can use to ease into these discovery conversations or design their relationships better.

Jean: Embrace that it’s going to be different from what you’re doing day-to-day. A lot of times people read communication books and they understand intellectually that this could work. Then, they go to work and they don’t use any of the things the book says to do because you don’t get the stepping stone of practicing in a safe environment and then transferring it to work.

Louis: As you both look forward, what’s next for Co Leadership? Are there any things you’d want Inside Intercom readers and listeners to know about?

Edmond: We’ve actually put together a guide for Inside Intercom listeners on how to design your relationships as alliances. It’s a tool that we put together to get a sample of some of the things that we teach in our leadership experiences as well as our leadership programs with companies. You’ll learn how to ease into a conversation where you actually have an explicit discovery phase of understanding what’s important to the other person, and then how to build alignment around that.

We’re trying to take a lot of these leadership experiences that we’ve been running in the Bay Area and broaden the reach to people all over the world. We’re starting to launch our online courses and build longer multi-week programs where people can learn some about investing in relationships, building trust, building effective teams, and we’re doing that across videos as well as experiential practice sessions that we do online.

Jean: We’re also running in-person events outside the Bay Area and doing some travel as well.

Louis: I’m super excited to see two people that are so competent attacking this for a lot of different companies and really helping them open their eyes to their potential. Thank you both for taking the time to talk with us. It’s been great.