When it comes to the importance of good people management, one data point speaks loudly above the rest. That is, people leave managers not companies.

Yet the transition from individual contributor to first-time manager is never easy, and there’s no perfect playbook. Engineering leadership coach Jean Hsu is on a mission to bridge that gap.

Following stints at Google, Pulse and most recently Medium, Jean works with technical leads through workshops and 1:1 coaching to help them get the most from their team beyond the immediate product cycle or feature release. She’s frequently writing about leadership on her blog, and just recently announced a new project with fellow engineering leader Edmond Lau, which you can learn more about here.

Recently, I hosted Jean on our podcast to talk about where she struggled in her own transition to management, the importance of psychological safety in the workplace, whether or not managers should keep their technical skills sharp, and more. If you enjoy the conversation, check out more episodes of our podcast. You can subscribe to it on iTunes or grab the RSS feed in your player of choice. Below is a lightly edited transcript of our chat.

Adam: Jean, welcome to Inside Intercom. Before becoming an Engineering Leadership Coach, you worked in-house as an IC and Engineering Lead at a range of tech companies. How did you get to where you are today?

Jean: I started as an intern at Google. That was my first and only tech internship and I converted to full-time, so I ended up moving to the Bay Area from New Jersey. I worked at Google Mountain View for about a year and half on the Checkout team and left with no plans other than wanting to see what else was out there. I did a bit of Android work on my own, and then ended up at Pulse, which is a news RSS reader. I worked on their Android app for almost exactly one year. At that point I was recruited to the Obvious Corporation, which later became Medium. I worked on the first Medium prototype. That was really where I spent a lot of time as an IC, and then Tech Lead and then eventually Engineering Manager.

I was at Medium for almost five and a half years. It was five people on the product team when I joined; the whole company was 160 at its peak when I was there.

Adam: Was there an a-ha moment that drove you to then leave Medium and go into leadership coaching?

Jean: At the end of 2016 there were a lot of reflection points for me. I just turned 30, so, you start thinking about what you’re doing with your life, and it was the end of the year, which is also a common reflection point. And then I had a just hit my five year mark too, so it was all those three things together. I started to think about what was next.

The role of a leadership coach

Adam: In practice, what does the role of an Engineering Leadership Coach look like? Because unless you have a coach, it’s hard to envision what coaching is.

Jean: What I do is usually a mix of coaching and mentorship, depending on the client. But coaching by itself is really helping someone figure out what it is they already know. That might be helping them get in touch more with what their inner critic says. What are the things that hold them back? And what is it that they want? A lot of people actually don’t ask that a lot, what do you want? Then you help them come up with a plan to get to what they want.

What are the things that hold them back? What is it that they want?

Adam: Does everyone have an answer right away for that though? What do you want?

Jean: You’re basically never asked it directly so, a lot of people don’t. Coaching helps them reflect back on what seems to excite them. So I might say something like, “When you talked about this project your face really lit up,” and then help them figure out what it is they get excited about so that they can then figure out what they want.

Adam: Is this a void that you felt personally in your own transition into leadership? Take me back to that transition. What stories or lessons stick with you?

Jean: I think what led me to coaching was realizing that it was a pretty rough transition, despite all the support I had. I had a really supportive manager. I had a lot of peers that I could lean on. And as an IC your calendar is pretty much open. I remember having switched to management and waking up one morning, looking at my calendar and I just had meetings from nine to five.

I was definitely in the mindset of, this is not the real work. I remember rolling out of bed, getting my laptop out, opening three pull requests and deleting a few hundred lines of code that I had just found within 10 minutes. And I was like, “Whew! Now I’ve done some work today, now I can go to my meetings.”

Now it’s clear to me that I wasn’t in the mindset yet that the meetings were actually the work, not just something that was happening to me and taking up my time. That’s a really hard transition to make because engineers spend so much time working with code, working with software building systems and then if they switch to tech lead or manager, they’re really expected to step away from that in some capacity. That can make dealing with your own self worth and productivity very difficult, because it can take a lot longer for some of the things in management to pan out and give you that feedback.

The management learning curve

Adam: You mentioned that you felt like you had pretty strong support at Medium, but is that is common in this industry? Or do a lot of the people that you coach with tend to not have access to that support?

Jean: I don’t think it’s common. Medium had a very unusually people-centric and supportive culture. When I first transitioned to management, the title was actually this term called Group Lead, and it was more of a coach really, because I didn’t work with a lot of the people that I was Group Lead for. So I came to management very much from the perspective of having to learn how to be a coach, I couldn’t just say, “Thanks for the feedback. I’ll go fix your problem now,” which you can do if you’re a tech lead manager. You can kind of lean into the, I’ll just go fix it for you, or I’ll just go do it myself.

Adam: But you have to some restraint there, right? Or it’s going to be harder for people to learn from those mistakes. Was it hard for you to step away from the code base?

Jean: It was hard to step away. Deleting code is very therapeutic for me, and there was this ongoing joke where my manager would say, “Oh, you seem kind of stressed out lately”, because he’d see five to 10 pull requests of code deletion. So, it was hard to step away.

I had this realization that it doesn’t do anyone good for me to clean up after them.

I had this realization one day that it doesn’t do anyone good for me to clean up after them. It was actually teaching them that someone is going to clean up after you, and I was depriving them of the joy of doing that. Because I had found it so joyful, I was really taking it away.

Adam: In the time since your own transition, whether it’s a first-time manager or a veteran in that role, are leaders struggling in a lot of those same places, or has that landscape evolved a little bit?

Jean: It depends on the person, but I definitely see a lot of common themes. For first-time team lead, or tech lead, there’s definitely that productivity issue when you’re stepping away from the work – a feeling of, “I don’t want to be to far away from it, because then what’s my value?” Then there’s also a common transition away from being the good student, the person who does the stuff that someone hands to you but isn’t really responsible for the decisions, you just do the stuff. There’s a transition from that to being the responsible party – the person who sees the problem, comes in with a solution and takes initiative, and rallies the troops.

Three workplaces, three cultures

Adam: Google, Pulse and Medium are very different in terms of size and scope. What’s something you took from the culture of each of those places that has just really stuck with you and helped shape the way that you would advise others?

Jean: Google really showed me what’s necessary at scale. So, I went into Google with no real sense of what it means to be a software engineer, and I came out with a much better sense of code reviews, technical rigor, all that stuff. That’s less important when you’re a startup of two people, but much more important when there’s 10,000 engineers working on one code base. It didn’t hit me that was because of scale until I moved to Pulse, and I realized that there were very few tests in the code base, there was a lot of copy paste, there was a lot of same file, inconsistent indentation, etc. It was very surprising to me.

But, I realized that they had done something at Pulse that I don’t think any team at Google would have been able to do. They shipped an iPhone, iPad, and Android app all within six months. That really showed me when do to throw things out and see what sticks, and when do you build it, because Google you don’t get to do that. Anything you ship is going to have millions of users on day one. There’s a level of, you have to build it right, even if it gets deprecated in a year. Whereas Pulse came from a very design school type of, see what happens, get a lot of user feedback, and iterate, and no one really knows us anyways, so we’ll just try a bunch of things and see what happens.

Medium I think is where I moved into management, but also saw the more people-centric side of management. I probably had that at Pulse and Google to some degree, but I wasn’t paying attention to it because I wasn’t in that role. At Medium I really felt like I moved more into finding what it meant to be a leader, and how to motivate others as well without doing more work.

Adam: How do you define people-centric management?

Jean: A lot of times people think we don’t have time to care about what people want or support people. They think that’s a luxury for companies that have it more figured out. I think the opposite if true, even if you’re in the middle of a release or even if you’re a startup finding product/market fit, if you can figure out what people want, that can help you align what the company wants with what the people want, and that can actually help everyone move more quickly. I don’t see that happening a lot in the industry and I feel like there’s kind of a dehumanizing aspect to that as well.

Adam: At Medium, you managed fairly large teams, with as many as 15 direct reports. That’s a lot, obviously. How were you able to make sure everyone got proper support?

Jean: We had an interesting structure. Sometimes I was running a product team with about 12 engineers, but I was managing maybe three quarters of them, and then also had four or five other direct reports. It was anywhere from 12 to 15 direct reports, and it was really difficult.

I know people who have 30-40 direct reports. That’s not management.

We had 1:1s every other week for 30 minutes, which I acknowledged was not enough, and I knew it wasn’t enough. I heard the analogy of circus plates spinning. You just kind of try and keep them up, and sometimes you’re like, “Ah I can’t do it. You take it”, or “Hire another plate spinner!” There were times when I had to be super upfront with my manager, and say, “This team is my focus, and I’m doing the 1:1s with these four people, but I’m not going in the meetings and if something really dire happens, I’ll let you know but that’s basically I’m at capacity.” Just being really upfront is important. He could have said, “Oh, actually this other team is more important.” So, just having really honest conversations with him was important.

Adam: Is there an ideal ratio for direct reports, or is it too circumstantial to say?

Jean: 15 was really a lot. I had this idea that if I had like six to eight people, then I could support them really well, but also have time and space to think about some larger strategic things. But I know people at Google or some other companies who have like 30 or 40 direct reports. And that’s not management. I mean, that’s something else right? That’s like having a reporting structure.

Adam: A lot is written about moving into management without, at least in terms of engineers without losing your technical edge. Does that really matter in a position where your impact is no longer measured in code commits? It is important is it to stay on top of those things?

Jean: There was this post recently about the IC manager pendulum, and that’s pretty common, to do IC for a few years, do management for a few years and go back and forth. Life changes, people change and you might want different things.
There are people who know for sure they always want to keep one foot in the technical work, but I think what’s important is to question why. When I was moving more fully into management, I remember talking to my manager and saying, “I feel like if I go anywhere else, people will not take me seriously, technically.” He prompted me to think about well, what would you get out of it? What are the skills that you would get out of going back and building more technical depth? A lot of it was just my inner critic of, well, maybe people won’t think I’m good enough. So, try to figure out if it’s because you truly want it, or is because you have self doubt about whether people will take you seriously or not. Imposter Syndrome. But there’s not a right or a wrong. Some people will always want to be coding at least part of the time.

Creating a psychologically safe work environment

Adam: One thing that you’re particularly passionate about is a culture of psychological safety. Just to set the groundwork for the topic, what does a psychologically safe environment versus unsafe environment look like?

Jean: For me a psychologically safe environment is one where people feel comfortable taking risks. They feel comfortable failing. They feel comfortable having really honest conversations and being vulnerable with the people around them. Even if you’re in a leadership position, open yourself up, have people see that you make mistakes and that you’re human. That can really set the stage for what goes on in the layers below a leader.

A psychologically unsafe environment is not super obvious, but one of the things I’ve seen is people aren’t taking initiative. That’s pretty common, where people tend to say, “We need hire more mission aligned people or, go-getters.” And it’s like, well, what about the people you did hire? A lot of the times it’s an environment where people don’t feel comfortable or safe stepping up and saying, “Hey I have this idea”, because maybe in the past they’ve been shut down or criticized without being recognized for stepping up. You can also look at the level of conversations teammates are having. If you find that people are only talking about work or people show up and they’re always like, “Things are fine”, and there’s not much depth to the conversation, that can also be an indicator of psychologically unsafe.

Even if you’re in a leadership position, open yourself up, have people see that you make mistakes and that you’re human.

Adam: So how do things tend to get this way? Because when it’s only the first handful of people in a room, they’re all questioning everything and have such great rapport. Is it a lack of onboarding for new people? I’m sure it’s gradual, but how do things come to a state where people feel like they can’t question the status quo in a working environment?

Jean: It’s a bunch of factors. Some of it is where people come from. So, if people have come from a more toxic workplace, they may just assume that’s the way things are, or maybe you add in a layer of management and people suddenly assume, “Oh, those are the people who are making the decisions and now I just execute.” And when those changes happen, not having those conversations around what are expectations, and what are the things you can bring to the table. I’ve seen that a lot.

Adam: As startups grow, hierarchy is one thing that ultimately has to be introduced, so does this come down to being more proactive in terms of communicating how an additional structure within the org is actually going to change, or maybe not change, the way people work?

Jean: Yeah, it’s being super clear about what it means to have a layer of management. Some people are used to management being like, well this person is just going to decide what I do. Every interaction with them is them judging me and jotting down notes for the next performance review. And I think a lot of people haven’t experienced management as, “Hey, I’m here to help you”, the servant-leadership aspect of it. Being super clear and building that trust early on is really important because otherwise people just make assumptions about what their managers are there for.

Adam: For anyone in management that might be listening to this episode, whether it’s something formal, like a questionnaire or a group activity, or maybe just informal questions they should be asking themselves, what can they do to take stock of where their team is on this spectrum of psychological safety.

Jean: I know there’s a few different types of surveys that you can take, but if it’s really bad, people are not gonna feel safe to talk about it.

Adam: So how do you course correct? Are these problems fixable?

Jean: It requires pretty upfront and honest conversations. I think it would be hard to admit that things are not going well. So, there’s a level of vulnerability in just saying, hey we took this survey and this does not look good, and let’s figure out what to do with that. But it really comes from leadership to set that example.

Finding your tiny victories

Adam: One of the best things about being an engineer at a startup is the feedback loops. And when you’re in the weeds as an IC you’re experiencing those all the time and when you move into a management role a lot of those things you’re working towards are more long term, quarterly, or annual company goals. How do you make sure you don’t lose sight of the small successes, and continue to celebrate them?

Jean: It can be difficult because the feedback loops are longer, and so sometimes you might think of the team execution in a quarter. How motivated are people, and how’s the team executing this quarter? That can be long enough that if you make any changes they can percolate a little bit. Some of it you kind of have to make up. I remember I had a direct report and he wasn’t that happy about the way projects were being assigned. It wasn’t even projects, it was more like tasks. And I don’t remember what I asked him but it was something around like, “What do you think should happen?” He then emailed the tech lead, and product manager and they completely changed the way it worked. They were like, “Oh, you don’t feel ownership, let’s change that. Let’s give you more visibility into what the problems are. And what the projects are that come from the problems, so you always have context on does this contribute to our purpose as a team.”

I felt really good about it, because I had that conversation right. But no one else knew, so if I didn’t really take credit for it, just internally, I don’t think any would have really given me credit. So you kind of have to be a little bit selfish and make up some stories about conversations you’ve had and the impact of them.

What’s next

Adam: Some exciting news just when public in your professional life, that you are going to be partnering with fellow engineering leader Edmund Lau in the coming year on a series of workshops. What are your plans for that?

Jean: We’re out to build the best leadership program for leaders in tech, starting with engineering leaders. Both of us have coached many engineering leaders, anywhere from tech leads, through to CTO’s. What we want to do is to take those learnings and put together workshops and frameworks, and move on to online courses so that those frameworks are accessible to anyone, not just people who can have access to us as individual coaches.

A lot of the workshops will be around having conversations as engineering leaders, really around the people-centric, trust-building aspect of leadership. And in the middle of the year, we’ll start to do some more online courses. We’re both in the Bay Area, so we hope to move from more onsite workshops and going into companies and doing workshops to more online and global work.

Adam: Five years from now, what kind of change are you hoping to have helped driven?

Jean: My dream is that everyone who works at a tech company will feel supported. They’ll feel like their managers care about them as humans, and they will feel connections with their manager and the people around them. That is something that is really lacking right now in the industry. There’s so much focus on the shipping and the product, and yes we work on and we work with software, but we’re humans. I want to saturate the market with people-centric managers so that when someone says, “I don’t feel that supported by my manager”, all their friends say they feel super supported by theirs. Then if that manager will then get feedback.

Right now I feel like if someone has trouble with their manager, it’s sort of tolerated. Companies have different reputations for how supportive management is. But it’s honestly few and far between. And the managers who are people-centric are few and far between as well.

Adam: Jean, thank you so much for joining us on the show today. It’s been a lot of fun.

Jean: Thanks.