Christian McCarrick  [0:00]

This is simple leadership. Welcome.

Thank you to our sponsor Auth0 for helping make the internet a safer place by offering identity as a service and for supporting this podcast.

We’re here to learn from new and seasoned technology leaders who all share a passion for improving the craft of technology management. Let’s take a deep dive into management and leadership challenges and best practices specific to Software Engineering and Technology teams. Do you want more engineering management leadership tactics and information? Subscribe at simpleleadership.io to receive the latest updates from this podcast. Hi, I’m your host Christian McCarrick. This is the simple leadership podcast. Welcome back. Today’s guest is Jocelyn Goldfein. Jocelyn is an American technology executive and investor. She’s the managing director and general partner at venture capital firm, Zetta Venture Partners. Previously, she was a director of engineering at Facebook and Vice President of Engineering at VMware. On today’s show, we discuss culture, hiring, seeing things from an investor point of view and supporting women leaders in tech

Good afternoon Jocelyn. Welcome to the show.

Jocelyn Goldfein  [1:02]

Thanks for having me.

Christian McCarrick  [1:03]

Absolutely. And my one of my favorite things is obviously recording live with my guests. And we are in the we work at Montgomery station here in downtown San Francisco. So thank you for kind of making the effort and coming coming in here as well.

Jocelyn Goldfein  [1:14]

Well, it was a very long journey from my office, three blocks away in Jackson Square.

Christian McCarrick  [1:19]

Excellent. Which if you’re in San Francisco, one of my favorite areas on an anecdote reminds me a little bit of Boston. I went to school outside of Boston, the brick buildings and everything. Yeah,

Jocelyn Goldfein  [1:27]

Yeah, I feel it’s like old San Francisco. It’s really got the personality of the city. And it’s actually got some of the historic buildings that have survived even the fires in 1906. Yes, it’s a really cool part of the city.

Christian McCarrick  [1:39]

Excellent. So for those who are not from San Francisco next time you visit Jackson Square should be on the stop of your tourist destinations.

Jocelyn Goldfein  [1:45]

Absolutely.

Christian McCarrick  [1:46]

Okay. Justin, like I asked all my guests, a little bit of the background like how did you get to be where you are today because it is an interesting one.

Jocelyn Goldfein  [1:52]

Definitely some twists and turns. Probably like most of your guests. I spent most of my career as a software engineer and then a technology leader and it’s kind of quite a late career left turn to find myself here in venture capital.

Studied computer science at Stanford, interned at Netscape at the dawn of the web browser internet era, worked for a couple different startups in the late 90s, early 2000s, including co-founding a startup of my own, about 30 seconds before the funding window slam shut. So really not in the boom era, really in the bust era, when I started my company, and it did fine in the end, but I landed at VMware in 2003 was the tech lead of the device virtualization team. And that company, just when I joined was getting product market fit with the data center. So it was a rocket ship, we doubled headcount and revenue every year, five years in a row. And in that environment, if you can get your work done, you get handed twice as much work and so I rose through the ranks pretty rapidly took on more and more responsibility rose to the VP engineering level. And even when we moved to a business unit structure

 

became VMware first general manager of a business unit. In my case, the desktop business unit in 2010, VMware, you know, from my perspective had grown huge. It was about 10,000 employees then I had joined when it was a couple hundred. So I thought I’m not cut out for this big company thing. I’m going to go join a startup. And the most compelling founder I met was Mark Zuckerberg, and in 2010, I joined Facebook. And so it was a little under 2000 people at that point, and also growing quickly, not quite 100% every year. And so it had some scale challenges that I thought I could help with, actually. So for years at Facebook, that was really my exposure to modern AI and machine learning techniques. I think Facebook was an early innovator in that space. And my first big project was helping adopt machine learning for the newsfeed also worked with the photos team.

 

We got serious about computer vision at that time, and a number of the other product teams really helped a lot think through a lot of the sort of organizational scaling issues at Facebook, like how we hired an onboard engineers

 

culture. My last big gig at Facebook was running the mobile engineering team during our third and final pivot to mobile. And along the way, sort of my last year so at Facebook had started doing some personal angel investing. And I reached a point in my career, maybe it was a midlife crisis where but you know, all those years at VMware, it felt like a vertical learning curve. It felt like every year I was doing twice at my scope was twice as big, I was challenged twice as hard. I had to run so fast just to keep up with my job duties. And the first few years at Facebook were like that, too, because it was so you know, you couldn’t find two companies more different Facebook and VMware. They’re both as a part of the tech industry. And so that curve continued. And then the last year or so at Facebook, I really felt like I was working just as hard. The problems were sure difficult, but I didn’t have to learn new skills. I was applying skills I had. And so I had this sort of moment. At this point. I’m in my late 30s. And I had this kind of moment looking in the mirror thinking

 

Hmm, maybe my 20s and 30s were for growth and my 40s and 50s will be for applying what I have learned all along, but maybe I’m maybe my growth has got to taper now. And,

 

you know, not in terms of seeking the next title or promotion, but in terms of seeking more skills than I already had. And then I realized, no, it’s just like, this is what happens when you do the same job for 15 years. And you could go down one path, which is just total mastery, perfection, I don’t want to say by any means that I was the perfect technology executive or, you know, got nothing wrong or had nothing to learn. But I think that I had definitely done the 20% of the job that was the 80% of the skill set.

 

Christian McCarrick  [5:41]

Sure.

 

Jocelyn Goldfein  [5:42]

And I felt well equipped to tackle any tech executive job. And so I thought, well, I could spend the rest of my life really mastering that long tail last 20% or I could find a completely new job again, yeah, and I and I just instantly knew that that was what I had to do.

 

And so I really at first thought that that meant starting a company and becoming a CEO.

 

Christian McCarrick  [6:05]

Okay.

 

Jocelyn Goldfein  [6:06]

But having previously started a company in the late 90s, you know, I can be honest and say we started that company because we wanted to be founders because we wanted to be entrepreneurs. And in hindsight, you know, I don’t think that’s a great motivation to start a company. It’s like having kids to save your marriage instead of for the sake of having kids.

 

Christian McCarrick  [6:20]

Yeah.

 

Jocelyn Goldfein  [6:21]

And so I didn’t want to create a company for the sake of creating a role for myself, I wanted to only found a company once I had a mission I was really enthused about. So I thought, well, if I hang out a shingle as an angel investor, I’ll be in the mix with other founders with ideas I’ll be exposed you know, here at Facebook, I’ve got my head down every day I can’t think and so and so I did. I just left Facebook and I you know, composed a goodbye email saying I was off to be an angel investor and, and and I started angel investing and I spent a couple years doing that. But the more investments I made, the more I realized that the work I was loving what was giving me joy was working across a lot of founders was helping a lot of founders.

 

Christian McCarrick  [7:00]

Sure.

 

Jocelyn Goldfein  [7:02]

And I thought actually, I want this to be my work. And I thought about just continuing as an angel. I explored running a syndicate through Angel list and a few other ways, but I really wanted to be significant capital, significant backing, for founders and significantly involved in and really helping get lift for the company. So that led me very long story short into into venture capital and, and through a series of hard work and lots of time spent looking, you know, was was able to be in the right place at the right time.. And meet the team at Zetta and joined forces.

 

Christian McCarrick  [7:34]

Great. You do point out something else, right? There’s is a huge difference probably between although they are both fast growing companies, VMware and Facebook, right. Do you think they were mostly different because of the time frame like when they were created and when they launched or just the market, they were going after enterprise versus consumer?

 

Jocelyn Goldfein  [7:52]

I think that the fundamental differences between the companies I think it’s really Darwinian, it’s really evolutionary. I think we

 

adapt and evolve to suit our environment.

 

Christian McCarrick  [8:02]

Sure.

 

Jocelyn Goldfein  [8:02]

So, yeah, the founding teams were very different. You can find more different founders than Diane Greene and Mark Zuckerberg, but also, Mark couldn’t have started VMware and Diane couldn’t have started Facebook, right? Like, it’s like they were well suited to the problems they were tackling. And so yeah, absolutely. I think that the core axes of difference were the market they were in, and the technology stack  that they were built on top of them. And the constraints that that created on release process, which I think is synonymous with culture.

 

Christian McCarrick  [8:31]

Yeah. Yeah, absolutely. Now going back to early days at VMware, or whenever you started becoming the manager for the first time, having gone through lots of growth at VMware, and then a Facebook and then where you are now, do you look back at your former self and say, Oh, my, if you could go back in time, what were some of the things you would say to some of the people who are like listening to this podcast now that might be more early stage managers that are kind of getting into are people that are thinking about going into management?

 

Jocelyn Goldfein  [8:58]

Well, I actually started managing

 

At the startups before that I worked for before VMware. And it was very much the battle probably I think a lot of engineers have this experience. It’s the battlefield promotion, it’s sort of look left look right. The team so big, we need a manager and who’s going to be it? And lo and behold your it. And so at the beginning, yes. Oh, my just about describes it.

 

Since then, you know, I’ve obviously grown in my craft, as a manager, I’ve transitioned a lot of engineers into their first manager role and train and supported them. I’ve tried to give what I didn’t have, I didn’t have somebody to learn from really, because I was in this chaotic startup environment where my boss was effectively someone with no more management experience than me, or no more engineering experience than me or both. And I had the gift in the early days at my own startup, of a co-founder who was really wonderful and was sort of figuring it out side by side with me, he’s actually a dear friend, and you know, he’s in Texas, and I’m in California, but we still talk every month and he actually gave me a great pep talk earlier this week, and I’m

 

reminded of how much we learned together and how much I learned from him. You know,

 

what I always say when I transition engineers, to managers is I cannot teach you the motivation. You know, I think a lot of engineers fail at that transition. Because at the end of the day, you know, they spend all day in meetings and they feel empty at the end of the day, like they’ve had no impact. They haven’t created anything, you know, if your internal motivation is all to build, to accomplish something tangible, to satisfy your curiosity, you will struggle in the management role. But if you really care about your team, if coaching somebody else and watching them develop or seeing the team as a whole, accomplished something, ship something, even if you didn’t write a line of code, but you could feel some sense of proprietary responsibility because you have enabled it. If that intrinsically, you know, lights you up, then I can teach you all the skills. And the skills, our communication skills, their alignment skills, their conflict skills, you know, you name it, but they’re all just skills and you can learn them

 

and there is, you know, even books about them that you can read. And there’s communities you can join and wonderful podcasts you can listen to. But you have to start by caring otherwise, it is a thankless job. And if your motivation to become a manager is to get authority, control, ego gratification, or access to, you know, privy information, like those are very common motivations to become a manager, and they stink, they will not guide you towards doing the hard work that actually makes you good at the job.

 

Christian McCarrick  [11:02]

It’s really anti patterns for good management.

 

Jocelyn Goldfein  [11:15]

Yeah, I mean, just think about if you wouldn’t want that manager, don’t be that manager.

 

Christian McCarrick  [11:34]

Yeah. And I often say to sometimes the most successful managers are the ones that are more the reluctant manager, right? They’re the ones that well, they weren’t quite looking for it. But like I said, people just kind of go towards you or you become the de facto person for that and, and now you’re a manager. And a lot of times those people become more of that servant leader type manager to the ones that are what you talked about, like they want the power to control the the authoritarians, right? That’s right. Exactly.

 

You mentioned something to and you posted this on a tweet, I think that what is it? Is it motivation? is one of the secret under utilized management tricks or something?

 

Jocelyn Goldfein  [12:08]

Superpowers!

 

Christian McCarrick  [12:09]

Superpowers, yes, yes, yes. Tell me about that. Like, why is that so important?

 

Jocelyn Goldfein  [12:13]

You know, as engineers, we are actually used to problem solving. We’re used to identifying a root cause of a problem and proposing a solution. And, and we think of ourselves as rational. And, and it’s, it’s actually, you know, I think engineers have a lot to offer the discipline of people management because we know how to think about systems problems. We know how to think about inputs and outputs and side effects. And I think in some ways, organizational problems are usually systems problems. And I’ve found that, that my engineering lens, I can sometimes see solutions to these problems, that other kinds of managers come up with different solutions. And so having that diversity of perspective is actually really valuable. And I think very little of the kind of business school pedagogy about management comes from engineers so I think there’s like some books to be written there. But as an engineer, sometimes we lose sight of the fact that

 

And we’re not fully rational actors, we may be predictably irrational, but we are, you know, we don’t always behave the way you want us to behave or the way and but people do respond to incentives. And the thing is the percent incentives are not just financial. And in fact, it’s rarely the financial incentives that matter the most. And we gravitate towards doing more of the things that make us feel like we accomplish our goals and personified who we want to be and lived up to the expectations of people around us. And we do less of the things that hurt and feel bad and feel shameful. And so praise and encouragement, you know, is like sunshine to a plant like people just grow towards it. And so you know, when you think about ways to motivate people and like, explaining to them how important it is or how urgent it is or how much of the people need it or you know how glorious it will be your your bonuses attached to this like all those are

 

blunt instruments were giving people a reason to do something. But I think praise and encouragement people just grow towards they just crave. And it satisfies such a need people have. And it also engender long term loyalty in a way that other motivations don’t.

 

Christian McCarrick  [14:15]

And you talk about incentives before too. And I want to tie that into a blog post that you had written about culture, you sort of talking about, you know, culture is that what you you know,

 

incentivize right? and potentially disincentive?

 

Jocelyn Goldfein  [14:28]

Yep. The title of that post is “Culture is the behavior you reward and punish.”

 

Christian McCarrick  [14:32]

Correct. So let’s dive into that a little bit. What do you mean by that? Like, how do you reward things? How do you punish things go into a little bit of the the high level of that blog post you wrote.

 

Jocelyn Goldfein  [14:41]

Well, another common phrase people say is, you know, culture is what people do when no one’s looking. So it’s not, you know, people following an order. It’s people knowing what to do and they don’t have orders, you know, and how do people know what to do?

 

How do they evolve a value system? You know, and I think there’s this idea that that culture is just like this ethos or this mood.

 

Or these traits or qualities that people possess and we interview for those, you know, culture fit. And then if someone or fits our culture, they have those same traits, we bring them in. And then we all act according to those traits. And I think that’s just sort of silly because companies are pretty unique from one another. And humans are obviously portable between companies that have sure culture fit, like we can assimilate. And also culture can change over time as well. And so what is it that causes people really to behave in a certain way and to kind of go with the herd? And I think it’s that people want to feel good. They want to feel a sense of I, that they fulfill their identity, they want to feel that they fulfill their goals, and they want to feel like they’re doing the right thing. I mean, I think all of us, for the most part, think of ourselves as good people or not, nobody wants to be bad, but also we want to succeed. And so when people walk in the door, they look around, and they look at the behavior exhibited by the most successful people and it’s like, Okay, I’m going to do what those people do, I’m going to imprint,

 

Christian McCarrick  [15:01]

even if that’s different from say, What’s written on the wall,

 

Jocelyn Goldfein  [16:00]

oh, I think what’s written on the wall, like, you know, they say that like 90% of meaning in a conversation is conveyed in like body language and eye contact and like only 10% in the words that you said. And I think it’s the same with culture like, like less, we’re less than 10%, influenced by the poster on the wall. And we’re 90% influenced by what people actually do. And in fact, if the poster on the wall is contradicted by what successful powerful people in the company do, then we become very cynical about because of that mismatch, when they’re the same. We can actually get into the like cult like scenario where where people really embody and you know, bleed blue for their company.

 

Christian McCarrick  [16:35]

Patagonia.

 

Jocelyn Goldfein  [17:00]

bump you know, I’m going to try to succeed. Sure. But at the same time when I look around to figure out how to be successful and look at other people, I can’t judge their success by their compensation because I don’t know it. That’s right. So I’m going to try to copy the people who look successful based on title, scope of responsibility, possibly just swagger. Sure, like how much confidence they exude, like how much their culture carrier and get to sort of speak up in conversations influence influence. Yep. Yep. Like, like we cannot like humans are very highly attuned sensitive social instruments to sort of figuring out who’s got status. Yep. And so there’s all these written and unwritten status triggers. And we are desperately trying to be like the high status people and we’re desperately trying to avoid being low status people. And that will lead us to all kinds of things. So I think one of the lessons you should draw from this is like, be really super careful how you hand out overt status symbols, like a title or a promotion, because everybody else is going to emulate that. And if you think well, it

 

You know, the classic example is the brilliant jerk. This guy’s you know, our best engineer, he’s 10 times more productive. He solves the hardest problems, but he’s so crusty and mean to people. And we have a culture of being nice. But, you know, we really can’t deny him this promotion. He honestly, he’s a principal engineer, so we’re just going to do it. And you know,

 

we’re just doing it in spite of him, you know, violating this culture of niceness. So you elevate the brilliant jerk. Well, guess what the message you’re sending is that people who succeed in this company and get ahead are jerks. And so you’re actually encouraging other people and you’re in you’re supporting. And you might think, Oh, no, people, no, the poster on the wall says Be nice. So you know, they’ll just realize that he’s an exception. Yeah. And you think that but it people always attach more weight to the behavior than to what is said, Yep. And so so then they’ll just think, Oh, I can be really cynical about that. That nice thing is just lip service.

 

Christian McCarrick  [18:50]

Sure. Yeah. And one of the things we just started doing to at Auth0 too is building some of those culture and value things that we want to emphasize on into our career letters, things like

 

diversity and inclusion and things like span of responsibility and coaching and mentoring. Those are all things that as we look around for especially when you talk about senior IC’s who might not traditionally be, have thought about, well, I’m a senior technologist, but as you want that span of influence to grow with your organization, you want those people to emulate the coaches that you want to have mentoring and sharing and guiding and coaching because that levels up the entire org. Right,

 

Jocelyn Goldfein  [19:27]

You got it. I mean, I feel like there’s a pretty early divide in an engineer’s career when they decide if they want to lead or not, and it’s not do I want to manage or not sure like, you hit the point of scope where how big a problem you can solve your is maxed out by your typing speed. Yep, yes. And at that point, if you want to solve a bigger problem than that, and you want to make decisions that span more characters than you can type, then you’ve got to be able to influence other people to work with you.

 

Christian McCarrick  [19:53]

And you want those people to have the values that you want to have your company espouse.

 

Jocelyn Goldfein  [19:57]

Well, yeah, and at no point like when you were getting

 

Christian McCarrick  [20:00]

Your computer science degree or learning programming on your own, or however you got into this career. And probably your first several years in at no point did anybody say, oh, by the way, communication skills and teamwork skills are essential to your success, right? Like we’re told we have to be smart, and rational and objective and good designers and creative and builders. But we’re taught that this is fundamentally solo work. And it’s such a lie, right? Because if you want to do anything significant, what one person can accomplish on their own is pretty small. And so like maybe you can get sort of one leveling up in scope just by becoming better at what you do, maybe two, you know, step functions in scope. But very early in your career, if you’re going to go from being from rote work, to meaningful creative work, the crucial skills, actually, the soft skills that were taught to sustain or that at the very least, are neglected. I think for a lot of engineers that comes is like a pretty unwelcome surprise.

 

But you talked about earlier on about humans as system, right? And it’s true if you could have if you look at it like a machine and easy to instance, right? There’s only so much in here from one. But now if you’re horizontally scaling, just like teams, you can have that multiplier effect and actually now serve a much,

 

Jocelyn Goldfein  [21:11]

Much harder problem. Yeah, have much more impact. Oh, yeah. Oh, yeah. And it’s so powerful to be able to parallel lives that way and to be able to tackle larger problems and decompose them. But it means a commitment to building a skill set that you may have hoped not to, or that you didn’t, you didn’t enter this career thinking that you that your ability to succeed was dependent. On other on those Yes, exactly.

 

Christian McCarrick  [21:33]

Which is a great point, which is a whole nother podcast episode about things we should teach right to getting into, you know, CS degrees or other ways to be successful in our careers. Aside from being up the whiteboard really well and interviews, right? That’s a whole other thing, too. Speaking of interviews and hiring, you had written another series on you know, hiring. I’m still working on it. Yes. Excellent. One of the things that stood out was the screening process, right? And you talked to

 

How we think people in general are cutting too much from a percentage at this phone screen level, right? going to talk about percentages and why and you into this whole thing. So tell my listeners a little bit about, you know, what’s the genesis of it?

 

Jocelyn Goldfein  [22:12]

Yes, yes. So this all came about because I was at Facebook, and we were trying to really ramp up our hiring. Let’s see in 2011. So it’s towards the end of the year in 2011. And our hiring target for 2011 was 300. Engineers. We had failed, we only hired 250 and totally 250. Engineers.

 

until proven innocent. We want to be as selective as possible and hiring a bad engineer is so damaging and has so much impact on the team that like, we want to be sure we’re only hiring the excellent. And so you know, when in doubt, vote no. And so if you take people that you’ve trained to interview in that way about bringing people on full time, and then you ask them to start picking up some phone screens, well, you’re just naturally like a bias to say no at the screening stage.

 

Christian McCarrick  [24:25]

When you start the biases, like I’m it’s you said guilty until proven it is.

 

Jocelyn Goldfein  [24:29]

Yeah. Then what you get is, you know, look, if we could be certain that someone was a higher after a single 45 minute conversation, sure, we wouldn’t need to do this extensive interview process where we talked to them multiple times, right? And so nobody sure after 45 minutes and so that they default to know then like it’s no for everyone. And we had a lot of also existence proof that we were losing very high quality people at the screen stage because we had all these employee referrals that were going into the hopper and getting rejected were very high quality engineers could attest that this is an

 

at source. Yeah. Like, yeah, like they didn’t even make it to the on site. Right, God. And so Facebook on the one hand is sort of very firmly attached to its traditions and hiring, you know, is kind of third rail for most organizations. Because, you know, if you mess with it, everybody’s, in some sense, we can consensus agree on the hiring process we have, because it’s the process that hired all of us, therefore, it must be good. Right? And you know, and if we change it, we could hire a lot of bozos. That would be bad. So, or people we don’t want to work with. And so

 

Christian McCarrick  [25:35]

or people that are different from us, which is another issue.

 

Jocelyn Goldfein  [25:38]

Also scary. Yeah. So in a lot in any organization, it would be very difficult to tinker with the hiring process, especially in a way that says, Well, before when we hired you know, we were hiring five out of 100 candidates and now we’re going to hire 10 out of 100 candidates. Like that’s intrinsically scary, you’re lowering the bar,

 

because now we’re less selective. And in truth, no. What I’m doing is finding and fixing all the darn

 

rather than have like a big knockdown, drag out philosophical argument with people about change, you can say, Well, I’m gonna try a test. And it was almost, you know, at that time, I don’t know, I can’t speak for Facebook today. I’ve been gone five years. But, you know, at that time, it was almost holy like if someone was going to run a test, you couldn’t stop.

 

Christian McCarrick  [26:47]

Sure. That’s a good culture. Yeah.

 

Jocelyn Goldfein  [26:50]

Well, it really has been. And so I said, Well, I just decided to run a test. And so my test was, I took a bunch of people who failed their phone screen and I brought them on site anyway. And I mean, not like, blindly or randomly sure it wasn’t entirely fair test. But people were, I looked at the resume or I looked at the employee referral, or I looked at the contents of what was in the screening feedback, not just the bottom line decision. And I just put my thumb on the scales, and I brought them on site. Now on the one hand, like,

 

that’s ghastly. But on the other hand is not, I wasn’t extending a job offer to them. At worst, the downside was we’d waste the three hours that we spent in an onsite loop. And at best we would discover what I discovered, which is that the candidates I brought through had very close to the same pass rate, the same offer rate as the ones.

 

Yeah, so the phone screen like looked like it was working because it was taking the funnel and reducing it. Right. We went from 100 candidates to 27 candidates, obviously, it’s doing something effective. Yeah. But it turned out that it was wrong in both directions. Sure, right, because we still only made offers to a quarter the people

 

that it did bring, so it had plenty of false positives. But just because you have false positives doesn’t mean you don’t have false negatives. And so we had a lot of false negatives as well, the way that recruiters typically dealt with the situation that a good looking candidate got a no on the phone screen is they would do a second phone screen and try to get a yes with a different engine with a different engineer. Yeah, but not tell them like not build on the previous sponsorship, like a total do over Yeah, it’s like, if you had code that had a bug in it, you would instead of fixing the bug, you would completely rewrite the code. Right? Like it’s 100% more effort. Yeah, just to answer that one. You know, just to just to get out that, that one niggle of doubt.

 

Christian McCarrick  [28:35]

And it’s also important, I think if you’re going to do that, that you don’t introduce that bias from a one person already saying no, right?

 

Jocelyn Goldfein  [28:41]

So that was the argument for you was that it has to be a completely independent. And then if you get a split Yes, no. Then you get those people to talk to each other and argue about it. And like under some circumstances, the recruiters even went and got a third screen done. So it could be Best two out of three. And by the time you’ve done three phone screens that are each 45 minute our online interview was only four slots.

 

So what is the point of a screen anyway? It’s not optimizing for the quality of our employees. That’s what we do in the onsite interview, when we decide about job offers, all a phone screen is doing, it is literally just an efficiency slider. It literally exists so that we don’t spend four hours on someone we could eliminate with one hour. And so by the time you’re spending three hours to eliminate spending four hours with someone, you actually and so I just did the math on here’s the average amount of phone screen time we’re spending per candidate. And if we just loosened the threshold, if we just said, you know, what, if I think there’s even a one in four chance that really matter if I think there’s even a one unfortunate so this person could get thrown on site. In other words, I think there’s a chance they could succeed. Not I think there’s a chance they could fail. Sure, then I’ll send them forward. And so what that means is we do more on sites, but we stopped doing these second and third phone screens. Okay, when in doubt, we send them through and so we actually were net neutral on time spent. Okay, yeah, with all the extra on sites. We made it up in the

 

It was really embarrassing and awful to have a false positive to send through a candidate that people you respect have to waste their time on. And unanimously vote no. And you feel shame and guilt, right? Whereas, at Facebook because we don’t hire for our own team, we do a central we have a centralized hiring process in which engineers are sort of phone screen any candidate. And then when the candidates are hired, they go into Facebook boot camp, they don’t even pick a team until

 

Christian McCarrick  [31:23]

Google, I think, has a similar process, right?

 

Google

 

Jocelyn Goldfein  [31:24]

Google has a similar process. But Google at least last time I paid close attention. Google just assigned you to a team. Whereas at Facebook, you came into the onboarding program first. And at the end of six weeks, the candidate chooses the team, like the team is all common for you, and you pick the team you want. So it was a good system, but it had this flaw, which is that you know, unlike in a world where I’m an engineer on a team, and I’m screening people who are my own potential future co-workers, and they’re the reinforcements they’re the cavalry. Like I have a motivation to add to my team, because I’m because I’m overworked. And come help me I faced with that motivation didn’t exist. Sure, right, because it was very indirect

 

So you’re not interviewing a candidate for your own team. And on the one hand that kept you sort of objective and didn’t make you desperate and willing to lower the bar. But on the other hand, it gave you no positive incentive to take a flyer on somebody, you just had this sort of negative incentive that I’ll be embarrassed. If someone comes through that’s rejected. So we just had spiral, those bad incentives could have kept compounding and would spiral and spiraled into this place where the screen was totally ineffective. And so we fixed it. And we’re a ton more productive.  

 

Christian McCarrick  [32:29]

sort

 

Sort of general philosophy concept, when you talk about the two different types of interview like Facebook, and Google during sort of an interview panel that is sort of independent versus manager, hiring manager, actually, your team members trying to hire their team. You also work with other companies now, what are the pros and cons of doing it one way or the other? And do you actually recommend one way or the other now?

 

Jocelyn Goldfein  [32:51]

Yeah, I mean, VMware was very much the classic model where a hiring manager hires for their own team. So I’ll call them centralized and decentralized.

 

Campus recruiting tends to be centralized even at companies that are otherwise decentralized. So the advantages of the decentralized model, every hiring manager for themselves is the hiring buck stops with somebody like the hiring manager has the right alignment of incentives to work hard to spend time hiring, because they get the benefit of it. And to make a really thoughtful, nuanced decision about who’s going to work with you, and also to bond with a candidate and can persuade them to come work for you to help sell right to sell to close, like Don’t forget, when we’re interviewing, it’s a two way street we got close to and so it’s better for making hitting your targets. And it’s also it’s probably better for fit. And it’s mostly just better for your accountability. And it’s better for people who do the interviewing work to feel that it has that it’s rewarded.

 

Unknown Speaker  [35:35]

actually that you can deploy your whole workforce. So hiring goals are distributed unevenly between teams. The teams who need to hire the most are this hit with the heaviest workload of hiring in a centralized model, you can you can actually load balance the teams who have the most time contributed and the teams who have the most need receive the higher sure, but a disadvantage. A really significant disadvantage of the centralized model is there is no hiring manager taking responsibility. And it both Facebook and

 

Jocelyn Goldfein  [36:00]

there has been a tendency to make recruiting the responsibility of the recruiter rather than of the engineer, which I think is toxic. Sure. Okay. What I actually love the most, and I’m not, I’m never a one size fits all advice person, I always think it depends on your situation in your company. But what I love the most is a blended model, which is largely centralized, but with particular hiring managers taking responsibility for a particular hiring target, and they go and they go work just as hard to hire that 10 engineers as they would if those engineers were joining their own teams. And you’ve got to structure responsibilities and culture and reward and praise and all the other things so that they feel as rewarded for that as they would if their team desperately needed 10 people got them.

 

Christian McCarrick  [36:41]

Sure. No great, great discussion on that. Shifting a bit now putting on your kind of venture capital hat. As you look into startups and other companies, is there any advice that you would give to engineering leaders at these companies that you just wish they paid more attention to, or wish they knew, or just something about could say, it’s not all about this, or technology or engineering, like pick your head up? Look around? Any advice for engineering leaders coming from the from the VC side?

 

Jocelyn Goldfein  [37:08]

It’s so funny, because I feel like actually, the commonest requests that I get from startup founders is can you help me hire my first head of engineering and, and, and that request comes about, of course, usually, because something’s going wrong. It just changes with phase and the birth of a startup, you’re not a company and you’re not trying to build for the ages, you’re just an engine of discovery, you’re just searching for product market fit. And that’s going to involve lots of backtracking, and you need as much feedback as quickly as possible. So as lousy as technical debt is, it’s generally by and large, it’s, it’s fine to accumulate a lot of technical debt early on, because honestly, the half life of the code you’re writing is a year or less, it’s going to be gone. So perfecting, it doesn’t matter if it was the wrong code to write if it was the wrong product to build. I think that this kind of the lean ethos has kind of gotten that message out there pretty thoroughly. But it’s still sort of easy to preach and hard to live. I do think though, there comes a point where you hit product market fit, and now you’re sprinting to satisfy as many customers as possible. And then what happens is you accumulate technical debt very quickly, and knowing when and how to sort of advocate for paying down technical debt. So you can speed up engineering, productivity is just an art. It’s totally situational. But But I do spend a lot of time counseling technical founders on like, when to make that call,

 

Christian McCarrick  [38:24]

and how to do it, because I think that’s something that engineering leaders, they just can’t say we’ve got a lot of technical debt, we have to pay it down, right? That that won’t fly at an executive team, right?

 

Jocelyn Goldfein  [38:32]

If the startup is 20 people and the CEO is also struggling, yeah. Or if it’s 50 people and the CEOs technical it can. But it’s engineers are prone to like technical, that’s just ugly to us when we want to fix it, because it’s an aesthetic problem, but also because, like it’s slowing us down. And it’s grungy, and it’s like working in a kitchen where every counter is covered with dirty dishes, like it’s just hard to work. But I think many CEOs have been burned with the engineer saying technical debt, technical debt, technical debt, and so then they take a quarter with no new features. And then at the end, the engineering team is no faster afterwards than before, because we rewrote the code, but we didn’t actually make ourselves more efficient. And so the other thing I would say is, the time to invest in technical debt is when you know you have the right product, and you’ve got to scale it. The other thing I would say is the time to invest in technical debt is when you actually have a solution that’s better, and for sure, will speed your team up afterwards. And you’ve got to be able to understand the cleanup, and explain the cleanup in terms of how you will go faster afterwards.

 

Christian McCarrick  [39:28]

Yep, what’s the ROI?

 

Another thing related to hiring related to culture, diversity and inclusion. I know that you also do work with female founders, women founders, and, you know, what are some things that CEOs can do startup founders can do engineering leaders can do to kind of really help embrace improving diversity inclusion in their organizations?

 

Jocelyn Goldfein  [39:51]

Inclusion is a function of your culture. Sure. And culture is created by the behaviors, your reward and punish. So guess what? people in positions change the culture are the people who are the role models and are and have the leavers of handing out status and rewards and punishments and firing people. So actually, CEOs and leaders are the only people and culture leaders, not just, you know, organizational leaders, those are the only people who can make the environment inclusive, like your diversity team can hardly do anything.

 

Christian McCarrick  [40:21]

Sure. The influencers? Yeah,

 

Jocelyn Goldfein  [40:23]

Yes. You know, there’s probably like a five step process of you know, the zero step is really assessing where you are, and whether your environment is inclusive. And if it’s not, what’s exclusive about it? And is that something you’re willing to change? Or is it not and just coming to grips with that. Or maybe there’s completely an inadvertent signals or something you had no desire interested in, you know, like this these few years back, I read this study, that’s like having Star Wars posters on the wall makes an environment, you know, less inviting two women. And I’m like, Oh, my God, that’s incredibly sad. I love Star Wars, when I see a Star Wars poster on the wall, like, I know, I’m among my tribe, and my people. Yeah, but like, that’s exactly it. It’s because it’s sending that clubhouse signal that makes you feel warm and belonging, but like, by definition, you feel belonging, because somebody else does not belong, right? There’s an in and an out group. And like, I don’t at the end of the day, I don’t need to tie my in group feelings of belonging to the movie Star Wars, much as it was the childhood favorite. And if doing that is keeping my team from being the best team it can be. And from hiring talent that I’m unable to hire right now then, like, I give up, I don’t need the Star Wars poster is not that important to me. So I think sort of things like sometimes there’s easy wins, you know, where it’s something that’s like symbolic and matters, but doesn’t really matter. And sometimes they’re tough. And you have to stick to your guns and say, you know, what, it is really important to us to have a no holds barred culture and that, like we allow swearing in the office, or we allow dogs in the office, right, this is a big divider. And you know, that means we’re not going to be inclusive to people with allergies, and we accept that trade off. Sure. Right. So you can decide. And then I think having made those decisions, you’ve just got to live them, and you’ve got a value, investing in diversity. I think when talking to engineers, I think there’s a subtext about diversity, which is, you know, I think all of us have the experience that we’re working at these high growth companies we’re desperately seeking to hire. we’re interviewing everybody, and we’re hiring everybody who’s above our bar. And so when we look at the result, and it’s only 5%, or 10%, female with the, you know, single digit percentages, black or Hispanic, some part of his is like, well, that must reflect the inputs. I know, I interviewed everybody who applied, you know, who had a good enough resume, and I know that I accepted, so like, to get a different population, I have to lower my bar, right, I’d have to accept people who right now are failing. And I understand that impulse. But I think it, it assumes a few things. First of all, it assumes that your interview bar is fair. And there’s like actually, a really data backed way to figure out if it is or not, which is to see how different populations make it through.

 

And  secondly, it assumes that the population who applies to work at your company is the population that could apply to work at your sharpening. And if you really value having a more diverse environment, and think it will help your company to have underrepresented people in the group to bring different perspectives and points of view, or to make it possible for you to hire more of them in the future, then you’ll go hunt them down. I mean, if you just sat there and only looked at applicants, you would never hire like that one signal processing engineer you needed. Or like, there’s always that one esoteric role, where it’s like, oh, I really need a network packet specialist and a networking specialist to if you know, who can do like, you know, Packet Inspection, and it’s like a very rarefied skill set, and it’s not there and your applicant pool, and you know, what you do you go source that candidate.

 

So you yeah, you’ve got to over rotate on finding candidates who bring things that your current team doesn’t have, and you’ve got to invest in that. And that does not require. So neither of those things requires lowering the bar.

 

Christian McCarrick  [43:58]

Just focus.

 

Jocelyn Goldfein  [43:59]

Yeah, just dedication. I think there’s also, I think, some tension around this idea that diverse teams perform better. So there’s lots of public evidence of this. There’s lots of studies know that public companies that have at least three female board members have better results, etc. Kinsey’s done a lot. But I think here in the tech industry, those are hard to take at face value. Because I think we can justifiably have some pride that our industry is one of the most successful industries of all industries, and that we have growth, that we have margin that we have, like the engine to make the world a better place for innovation, like

 

anything you could measure an industry on, you know, compared to, I don’t know, agriculture, politics, finance, whatever, like, I think we can justifiably feel some pride. And we’re also like, among the most images, yes. So if diversity is such a strength, why are we why is homogeneous tech better

 

than, actually those other industries are pretty homogeneous too,

 

but not as nice as us. Sure. So. And in particular, if we look at our most luminary companies, the most successful companies like remember culture is the behavior you reward and punish, homogeneity has been rewarded in the tech industry, it hasn’t been punished, like by performance. Sure. So this idea that we’re a meritocracy is like, a really powerful idea. And there’s the suspicion of hiring for any reason other than merit. And I feel that, actually, I kind of agree with that. And I think that most diversity advocacy doesn’t really come to grips with that reality. And so I think the principal position you can take about diversity and inclusion, and I do take it is that we would be even better sure if we were diverse. And moreover, I think tech, unlike other industries, is not a fixed pie. There’s just like a finite amount of jobs that can exist for lawyers, that is a function of total population size, there’s only so many lawsuits that can go on, there’s only so many contracts that can go on, it’s a function of how many entities exist. Same with doctors, right, like given a certain population of people who can be patients, we only need so many doctors and nurses. Tech is an unbounded pie. The more technically capable people we have, the more inventions we can create, the more value and wealth and impact we can have, the more tech workers we can absorb. So I would argue that even if you think tech is awesome, the way it is, if we could double the size of the tech industry and double the tech workforce,

 

to like who could argue with that?

 

Christian McCarrick  [46:20]

Sure. And the impact would be exponential

 

Jocelyn Goldfein  [46:23]

exponential. So whether or not you

 

think that like adding women and brown people to the organization makes it higher quality? Surely you can argue that like, okay, actually hitting my hiring target instead of half of it would be great. And by setting ourselves up in these exclusive ways, and by losing all that talent in the world, that could be exceptional tech workers, because of these social stigmas and stereotypes. Like that’s crazy. Like, we’re just we’re operating at half strength. Sure. So that’s been most of my belief most of the time is just that the argument is just for growth period. And like the obvious cavalry to tap, like, I think we found every white and Asian man who wants to be a tech worker as a tech worker by now. Okay. But like the obvious cap, the only place the cavalry is going to come from is we go tap into these people who are underrepresented. But I also in the last few years, with the rise of AI, and social media, the rise of the ethical questions around it, data privacy, I think all the social impact sort of knock on impact questions of what we’re building. I actually think for the first time I feel in my career, I feel like I can really point to a set of issues where I think if the leadership of tech industry were significantly more diverse, we would actually see these companies perform significantly better. I think think we’ve been blindsided and too slow, because we’ve been to homogenize at the top.

 

Christian McCarrick  [47:43]

Well, great insights. Thank you for sharing that. As we kind of get to wrapping up here, I asked all my guests, any recommendations you have, it could be leadership, Ai, books, podcasts, you know, anything that you have that you might recommend, for my listeners to, to absorbe that you think might help them?

 

Jocelyn Goldfein  [48:01]

Well, I recommend my blog, of course, in find it on Medium, my my spelling of my name is unique. I am a big fan of a book called “Influence” by Robert Chidini it’s not extensively about management, it’s very much about if you want to see humans as a systems problem, and you want to understand the inputs and outputs of human behavior. And you want to shed the delusion that humans are rational. This is a really good book.

 

Christian McCarrick  [48:27]

And we’ve pointed out they are not rational.

 

Jocelyn Goldfein  [48:30]

Well, I think we’re predetectable, but maybe not, it sort of depends on how you define rationality. Maybe I also really love a very old book from the 80s, which I think you can still find in a reprint on Amazon called “Debugging the Development Process” You know, there’s lots of books about management that talk about sort of the people and soft skills and coaching and their side of hiring whatever side of management, there’s very few that talk about the project management inside of our job is kind of dated, because it’s kind of back in the native software days. But you know, as much as it’s been written about Lean and Agile, I think this book is really unique and how it blends the people in project management responsibilities, and how those two things are actually inextricably intertwined. Like we don’t just have HR managers managing our teams, it has to be an engineer has to be somebody who can make an intelligent call about who should do what and whether they’ve done it correctly. Great. So that looks kind of unique.

 

Christian McCarrick  [49:25]

Okay. And for the listeners out there too you mentioned your blog, what are some of the best ways people could reach out to you if they wanted to contact you?

 

Jocelyn Goldfein  [49:33]

Well, I’m email is is probably the easiest way I’m Jocelyn at Zetta VP dot com. Okay, but you can follow me on Twitter, and LinkedIn connection. And yeah, I’m easy to find because there aren’t any other Jocelyn Goldfein’s in the world.

 

Christian McCarrick  [49:45]

Excellent. So I will put all those to listeners as well. I’ll put them on simpleleadership.io so you can actually see the links to these as well. Jocelyn, thank you very much for coming in today really had enjoyed my conversation.

 

Jocelyn Goldfein  [49:55]

I could talk all day about these topics.

 

Christian McCarrick  [49:57]

Okay, thank you very much.

 

Jocelyn Goldfein  [49:58]

Thank you.

 

Christian McCarrick  [49:59]

Thank you for listening this episode of the Simple Leadership podcast hosted by me Christian McCarrick. If you have enjoyed the show, please subscribe. And don’t forget to leave a review on iTunes. Full show notes and additional information can be found on simpleleadership.io. If you know someone who would be a great guest for this show, or you want to share your own experiences, please drop me a line. We’ll see you back next week for more technology leadership tips and advice as I interview more top software engineering leaders