Juni 24, 2019

Episode 7: Is Bottom Up Innovation Really Possible At Gojek?

Episode 7: Is Bottom Up Innovation Really Possible At Gojek?

Intro: Hey guys, welcome to GO FIGURE. My name is Nadiem Makarim, CEO and founder of GOJEK Southeast Asia's first Super App, which recently became a Decacorn GO FIGURE as a podcast dedicated to expose the inner workings of ambitious tech companies in the emerging world. Hope you enjoy it.

Nadiem: Hey guys, welcome to GO FIGURE. Very, very excited about this podcast. We got Ajey Gore in the house, our global CTO. Ajey, how are you? Thank you for coming by.

Ajey: Thank you for having me.

Nadiem: Ah, yes, thank you very much. And we have Sidu our Head of Global Talent and Head of India office, who has already made an appearance on our podcast as well. Um, today is a topic that is very close to my heart, a topic that I feel is very vague and very buzzworthy. And yet not many people have broken it down into more scientific terms. And we're going to talk today about culture. We're going to talk about company culture. This has been a buzz word. It keeps being used everywhere and around. But if I had to ask any average person in our company or outside about what it means to have a great company culture, everyone will spit out a different response. Right? Everyone has at some points, wildly different views about what constitutes a good culture and a bad culture. And so, I don't want to say good or bad in our discussions, but I want to define it as high performing cultures and low performing cultures. All right. Um, first of all, I want to kick off the conversation. Why is culture so important in an organization? What is it? What is it and why is it so important? How do we even begin to define this?

Sidu: So you know, just like you said, everyone's going to have a different answers. So caveat it up front. This is mine, right.

Nadiem: I only want standardize answers too. Boring, boring answer.

Ajey: Not going to happen, Bro.

Nadiem: All right, that's good Jack.

Sidu: So, so the way I like to think about it now, and this is something which has evolved over time, is in the beginning I honestly thought it was fluff. I thought this is, this is, I mean, I'm a programmer, right? I, I started as a programmer. I've been coding since I was a kid and I got, this is some management fluff that's nonsense. Uh, an overtime I started to realize that in an organization there are certain things that you want people to do and certain things you don't want them to do. And usually you try to define these through processes and policies, but those number of things that you can or cannot do it so fast that if you attempt to codify all of it into policy, either you get this massive book of rules and it's still incomplete, uh, which people don't grasp or follow. It's un-applicable or you create this horrible bureaucracy or a combination thereof. So for me, culture is everything that isn't process or policy, but that still defines what you do or don't do.

Nadiem: Definds what you do or how you do it?

Sidu: Or how you do it. You know, the choices you make every day.

Nadiem: But it has to be manifest in a it an activity. An action.

Sidu: Yeah, the results of it. These are decisions which people make several times a day, hundreds of times sometimes, and you can codify all of them. It's the stuff that you don't codify. What, what do people do for those decisions? That's the culture, right?

Nadiem: So culture, first of all, first rule of culture. Culture isn't something that is individual and kept inside your heart or your mind at all times. Culture is a outward manifestation of a shared set of values. And the only way to identify it is through action. What do we do? Yes. What do we do?

Ajey: Or you can say culture is a collective expression of a group. You can look at it the way it's, so basically whenever you looks say what is the culture of this country or this company or this place, what you define by, when you meet a bunch of people from that specific implementation, how do they express yourself? Right. And how do they go about expressing themselves within the organization? So you can actually quantify it or you can just say it like culture is collective expression of a group in a certain way and that's what really differentiates them with a different kind of organization.

Nadiem: So it has to be somewhat unique to that organization. Otherwise it's not a culture. It's something shared within that culture. Cultures can be similar between companies.

Ajey: Some traits can be similar,

Nadiem: but they're never the same, right?

Sidu: Yeah. I mean why do the trains run on time and some countries and why do they not run on time in other countries? Is it in the constitution? No, it isn't. Yet there are countries where the trains run on time and other countries where they don't.

Nadiem: So you're saying that trains running on time or no, is a manifestation of culture. Interesting.

Ajey: If you look now, let's put it on down to codify things, punctuality is a culture.

Nadiem: Is a culture. Yep. That's right.

Ajey: If you look at punctuality, like how much serious weight it's people leaders do to being punctual, being on time. If you start doing that repetitively, then punctuality becomes part of culture. It becomes that one codified DNA that this organisation is punctual.

Nadiem: Okay. So that's the second telltale, the first telltale, it has to manifest into action. It's not a thought or a feeling. It has to be a collective set of action. The second thing is what you just said is that it has to be repeated. Yes. Right? Consistent or relatively consistent. Right? And by the members that belong to that culture will generally behave in a certain way.

Ajey: Yup. For example, when there's not a consistent across and what happens is when it's majority consistent, the other people start following. So a lot of time the culture is something which gets codified in early days and over a period of time it kind of become rules. And then if you have this, uh, manifestation of challenging things, then culture cultivates and motivates people and also molds around the rate. So what our culture you have today, maybe GOJEK at the DNA level, will still have that culture down the line say 10 years, but the manifestation of expression of that culture will be very different. It can be there as well.

Nadiem: So it evolves. It can evolve. And in many ways it probably should evolve.

Ajey: It has to.

Nadiem: Right? It has to evolve because it's, there's no, it's not a set rule book. It's how the collective has decided to behave with one another.

Sidu: And it's very rarely that the culture is set by the minority. There are very narrow situations where a minority can set the culture for the majority. It's almost always the majority norm that defines the culture.

Nadiem: Hah. But here comes there. Interesting point. There is a paradigm or a set of beliefs that think that culture is very much top down but are not very much top down. But it begins at the top. We've heard this many times, culture begins at the top

Ajey: In early days.

Nadiem: Sure. Yeah.

Ajey: It later days it do not. So suppose I'll give you an example. Suppose we get somebody else heading a big organization and he wants to have or she wants to have their own way of working. It will not go down.

Sidu: Punctuality.

Ajey: Punctuality, for example. If everyone is lazy or everybody doesn't come on time. And then somebody one day a boss comes and we have to be doing on 9AM and nobody's going to show up. So it is early days also matters. So culture also depends on things like, so if you look at us, the way we grew our culture over a period of time in the last three and a half, four years. There is more about all of us collectively. It was not like that we open some books and we got to do this tomorrow. We did not do that. All of us had our own traits and also culture is kind of melting pot of the good values across all the leaders and people. So basically over the period of time, what we did, we kind of adopted nice things from everybody and kind of discarded things which did not work out. And that's what actually comes out as a culture. So basically if you look at, if you look at nature, right, it's survival of the fittest, right? Same way it is and the mutations, like if you look at grizzly bear versus um, what is that bear bear

Nadiem: Black bear?

Ajey: Black bear or polar bear, the polar bear. There was mutation happened with polar bear, they became white and that's why they are white, right? So culture is also like that you have this DNA mutations within the leaders within the people and then finally the best thing comes out.

Nadiem: So it's like a social Darwinism that behaves in it. So there's a self selection process that happens to it. So you're, you're talking about this as a much more organic kind of, you know, factor of evolution. Almost accidental. But when we look at the books, when we look at all of that business school academia, etc, this is something you have to harness, shape, direct like and how do you reconcile the topic? I can never deny that I have seen manifest in the organization multiple things that I've noticed our core leadership team doing. Like I can't deny that. Like there's definitely a role model influence. Doesn't, it doesn't mean that we define all the culture. No. But there are traits of the culture that I see that it's reflected through our top leaders.

Sidu: I have a favorite example, which I said it's frankly, it's one of the reasons why I work here, right? It's not that, you know, I'm spoiled, I'm not spoilt for choice, which is, and I noticed this in the early days, which is, you know, someone proposes a feature or a change to the product, to the APP that exploits the driver. And I've seen you, especially in the early days when, you know, we were all on the groups together, you know, discussing these things or debating these things are making these calls where you would jump on that person immediately and say, we will not do things which harm our drivers. We are not going to go zero sum on our drivers. We don't do that. And then over time I noticed this trend very clearly that there is this very clear bias to removing or eliminating choices that harm our drivers. Uh, you know, there is, there is no conversation that happens in a project that lasts for any length of time, which involves, hey, let's do this exploitative thing, which gives us an advantage on GMV or whatever. Pick a metric at the cost of a driver that's the lives for any length of time and you're no longer in those groups. So how did that happen? That's for me a great example and like I said, you know, it's one of the reasons I work here. I don't think that's a very common trait in, in ride sharing companies.

Ajey: So I'll give one more example. If you look at this, we recently like four quarters, no eight quarters back. We start looking at OKRs. Yeah. It was not dealt with us.

Nadiem: OKRs are Objectives and Key Results. Target setting.

Ajey: Yes. And we started looking at those as one of the ways we can operate ourselves and make sure that we are, we are rational across product roadmaps and we are rational across all the data and everything else and that is stuck on, that has stuck on because it was the right thing to do. And many, many, if you, if you think it's just top down, it will, it would have not been top down unless people saw merits in it. Right? So over the period of time what happens that the basic manifesto or the basic, what do you say, principles which we won't deviate from, are very simple. We will adopt, but things which are good for all of us. We will never go and do anything but she does a harm to our customers, merchants or drivers. I still remember in our early days when we used to suspend driver's or we used to do something with customers? You used to be so anal about it. Go find the reason why did you do this? How can you log out 500 drivers that important and the number was very small at that point of time. But we still take care of those people, right? So over the period of time, if you look at culture, culture is manifestation of good things, interruption for augmentation and then collectively expressing that as the right thing for everybody else

Nadiem: Without having it written.

Ajey: Without having it written.

Nadiem: And so culture is essentially a far more superior tool of management. It's not a tool management, it's uh, but it's, it's a far more efficient scalable way of creating shared behavior than any other thing. And this, I agree, there's no amount of like guide books, manuals, 10 core values, whatever it is. Those are things that can help scale up maybe the initial founding teams or the initial leadership teams notion of what good behavior is of what the correct behavior is. But if you don't codify behaviors culturally, then you will always have to define it through processes and policies, which are hugely inefficient,

Sidu: And just not scalable.

Nadiem: They're not scalable.

Ajey: I'll give you one more example. How many times did you let somebody go because it was not culture fit?

Nadiem: Most times. That is literally the number one reason why people are let go in GOJEK.

Ajey: How many times have you not able to codify and put in the templates that you did not check this box.

Nadiem: It's really hard.

Ajey: Correct. So the culture is a manifestation of expression. And when somebody does not express the way we want them to express or somebody does not accept our expressions, that's where the misfit of culture happens.

Nadiem: All right, let's, let's double click on that because that's a perfect way, maybe a back door way of describing what culture is. Let's talk about letting go of people because they don't fit our culture right. What's usually the reason, like the last three people even, and what are the top maybe three reasons? So when you ask, when someone says, Oh, uh, why, why did that person get let go? Oh, he's really wasn't a cultural fit when you ask them, okay, so what, what did he do that insurer, what are that usually?

Ajey: Usually because they've violated one of our values. Ethics, integrity. So that is one, one part of it. Second they've violated one of the principles where we actually wanted them to do something and they didn't do it. And not because we felt that it was wrong. But the team felt that is was wrong. So it's organization driven, it's team driven. And third thing, which always I have seen is that they may not be happy because they could not find a place in in our oraganization because they are used to doing those things. So they maybe come from a very different culture, which may not be wrong because that's the culture they grew up. And then a lot of companies out there who behave very differently.

Nadiem: That's right.

Ajey: That's what these are the three reasons.

Nadiem: So you don't necessarily need to be a bad manager or bad leader or a poor performer per se. You may encounter huge amounts of cultural friction. Right? This concept of cultural friction in an organization that is completely different, and I've seen this happen to a lot of senior hires that we have when they're coming in and they start, they come from a much more hierarchical culture, a much more, um, uh, almost like revering leadership culture, which is not GOJEK. And they start shooting out orders. They come in here and they start saying, okay, in a very structured way, even politely. Right. And you're just going like, oh, you should do this, this, this, you do that. You do that, you do that in a very shoot a shoot first, you know, ask questions later. A company culture and then they get frustrated because no one actually does it.

Ajey: And what happens? One more thing. Is their reporting line, which we don't have but whoever works with them to start reaching out to you or me or somebody else and they don't like it as well. But that also happens like how did you actually bypass me and go and our culture is being very transparent and very open. So our culture allows and let people operate in that manner and they are like, don't know why but they don't like it. So maybe from where they came it was a perfectly fine thing, but over here they're not a cultural fit. And if they can adopt there are a lot of leaders that we have got, they adopted very fast. They realize the value of being open. They realize the value of being transparent. They realize the value of being open. That means they can, if somebody goes and talks to us or somebody else, they can get more help. They don't see demeaning, they see it very useful.

Nadiem: But I think. I think it's important, Ajey to put a caveat to this by saying like not all of the elements of our company culture are necessarily the best. There's no such, it's hard to define. That's why again, there's no such thing as like bad culture, poor culture. There are, there are cul- but if you're in without the ability to adapt to an existing culture, you will almost inadvertently fail.

Ajey: Yeah. But there is something about poor culture.

Nadiem: Because for example, the hierarchical thing, right? The consequence of that cultural trait of ours is that it takes a longer time to get people to buy in to what you want to do and execute. You need to spend time explaining the why, the reason, how it fits to the bigger vision, the urgency and all of this stuff. It takes a lot longer for a leader to start executing once they come in to GOJEK, but once they start executing and everyone gets the buy in, they're rolling.

Sidu: For people who come in with that background, they're very used to having to align upwards but there unused to having to align sideways and downwards, which is an expectation here. Yeah. Right.

Nadiem: And in most, and to be honest in most Asian professional culture's, it's very

Sidu: alien.

Nadiem: yeah, exactly. It's a very alien to have to, why do I have to justify? I'm the leader. Why do I tell you the, you're, you're a young kid. I'm obviously hired to be above you and I have more experience than you

Ajey: but if you look at, if we look at one of the values, would you believe in is bottom up innovation? If we don't facilitate them that thing, then we would have never been where we are. We would've never allowed the bottom up innovation. We would have killed the ideas from ground up in infancy stage itself because they would have never reached anywhere.

Sidu: The way I like to think about it as you have effective and ineffective cultures which is in the context of what that organization is attempting to achieve.

Nadiem: Yes,

Sidu: An effective culture may not be necessarily pleasant for somebody coming in from another effective culture. Yes, it is not a commentary on whether a culture is right or wrong. Cultures are effective or ineffective given the goals of the organization, but it doesn't mean that one effective culture will translate well for someone familiar with it who moves to another effective culture.

Nadiem: Yes. It has very strange how there is almost like you take the analogy of a of a foreign invader or a pathogen inside a body, and I don't mean this in like a negative way, it's just it's, it's analogous. The mechanism. As soon as they're a very different culture base enters another culture base which is different to them. There is almost this automatic auto immune response. Have you noticed this? And I can't put my finger on how it happens, but people start coming up to me or you and say, "Hey, this guy, you know, how do I really deal with him?" You know, like or this guy or girl and there's this natural almost like it's not a rejection straight out, but it is creating inflammation. You'll immediately see that happening and that's when you know your culture is thick, when the body treats you as like a foreign pathogen coming into side. Right?

Ajey: So anytime a leader joins us, uh, especially in engineering or product management department and they asked me or anywhere else what and they reach out to me and say what is your first advice would you give me? So I said for the first four weeks, observe the next four weeks absorb. And then after that, and that is very important because you have to become one before you can execute like one.

Nadiem: Yes.

Ajey: And whoever... A lot of times people have come back to me, senior leadership has come back to me saying that was a really a nice advice you given. And then I told them if you would have just gone ahead and act on things the way you see on the surface, you will start assuming a lot of things. Once you assume anything, you always, assume one thing in negative connotation, you always think why this cup is here, why this cup is on the book and you look at him and assume Ajey is an idiot because he did it like that. But there is reason for it. If you ask me why, I can tell you right now why? Because it doesn't make a noise over here, right? Same way in our organization, there's a lot of the things. We are 36 months old or 48 months or we are like a baby elephant who just started walking and doesn't know where to look for and one of the biggest thing which we have learned over the period of time is that every time when we have done some mistake, there is a valid, very valid reason for it. When a leadership comes from outside and they see these five things the way they are done and they're absolutely wrong because it's not done somewhere else. If they don't observe and absorb, they will never realize the reason for it. And once they realize the reason for it. They can actually empathize and correct it in a much better way then doing the knee-jerk reaction.

Sidu: It's also why, uh, you know we almost as a matter of policy now expect leadership to spend time in the trenches executing for some percentage of their time for exactly this reason. It's why on the engineering side we have this rule that says that everyone must spend time coding irrespective of how senior they are abstracted away from the day to day stuff they are. You need to spend some percentage of your time writing code because otherwise you will not understand why certain decisions are being made by other people because it's not obvious at the surface and there is no scalable way to surface that.

Ajey: Even for our drivers, remember what we did. So when we did launch GO CAR, we made Kevin drive around. So you know one of the incidents that happened to Kevin was; a lady got him as a driver and then she took him to a laundromat and made him wait over there. She said it'll take two minutes I'll come back and she made him wait near the laundromat and he got like super frustrated like this is what our drivers go through? We need to fix this. And then later on what we did, we sent a bunch of our PM's on the road as a GO RIDE drivers and they were like getting super frustrated this app. Is this is what our drivers go through? And that kind of empathy and observe and absorb works really well. If it can work well there it has to work well in the organization as well. And that's what defines our culture because we treat everybody equally when they come here. We make sure that there's no question on the intent and don't assume anything. If you look at this, this is our culture. How do you define this? You can't codify it. It can be handed down from generation to generation in startups like us, generation is only six months to eight months because they're infusing new people all the time. So that's what happens and that's what kind of becomes culture. So people know we got to do right thing, we don't have to assume. We don't have to question people's intent. We gotta go ask why this happened? Instead of why are you not doing this? Those kind of things. And that is openness. Transparency. Like when you go and talk in our town halls and we are very transparent with what our numbers and what we are doing, we will get the confidence. Look, this is what we are when our leadership does not hide anything from us and why we should hide anything from them. And that's why, how many times we had downtimes right? We did have downtimes. How many times you went and talked to so many engineers and you've got the same answer. It's not a cause for concern. It's just being transparency is a culture. Because what we did, we put a psychological safety or there. That is part of culture. Now if you start talking about these things and then this 10x of culture will start coming out.

Nadiem: Yeah. I used to be super for.... So for the listeners, whenever we had any downtime, you know, I had, I came from kind of the conventional professional management, uh, background. And so I would triangulate things that happened because I made the assumption that there was something else people won't tell the leader or the boss what's really happening. And so at several times in GOJEK. I went to like three or four different people asking, okay, what happened? Really what happened really? And when I got the same response after the third one, I'm like, I felt really bad. I was like, okay, so they are telling the truth. I was like um...

Sidu: And this isn't to say that in that particular example, this isn't to say that there aren't effective cultures that, uh, you know, do not need this kind of hands on coding for example. Right. There are very effective cultures that function where leadership does not code, but we have an effective culture where leader do code.

Nadiem: Yeah. And, and it's again, like your point that you said it depends on what the goal is. Yeah. And I believe we've hit this critical point whereby the complexity of our business is at such a high level, um, that death and the very definition of a performance culture needs to be dramatically different because we've hit that point where by leaders trying to exert the more our top leaders try to absorb decision making, the worst outcomes we will get. Because we're so far away from the actual problem at this stage. So, so along this lines, how do we then define what a high performing team is for a, you know, like us top 20 private company in terms of valuation in the world tech company at that size, what defines and at that level of interdependency, that level of complexity, what defines a high performing culture? And if we can define what that culture is, why don't we break it down one level down and say, what are the high performing beliefs or behavior that we believe can take us and be able to make sense of this complexity and succeed as a business?

Ajey: Actually look at our values.

Nadiem: Well that's what values are. Values are our core values are a reflection or an attempt. The good or bad, I don't know, but an attempt, an effort to codify what we believe to be the positive elements of our culture.

Ajey: So in that sense, if we look at one of the biggest, uh, two two of them is walk the talk. We want our leaders to talk, right? Second, we want them to earn the title. These are two things. That means

Nadiem: Earn your title, that's one of our core values.

Ajey: So if they want to earn it on title and if they're going to walk the talk, that means they should be practitioners, they should be craftsman. They should be making and refining their skills all the time. If you look around, all of our leaders are actually doing that. They are in the trenches of people. They are the role models for their departments. And that's why it defines that way. If you, and that's why we are the very, that that's why we are so small team doing operating such a large business, right? If you don't do that, if I start putting my walk the talk uh, skills aside and starting making sure I have proxies to do all the talk and they have their proxies to walk the talk, so we'll end up with a such a big large team because I'm not doing my work and there are a lot of companies out there with our valuation and our size, which has six times the teams, engineering teams that like we are, we are only three hundred people right. There are people out there comparatively, uh, companies out there who are 2,200, 2,500, 3000 people. Why? The reason is that because those people believe or those people figured out that they can hire somebody else who to walk their talk and then they start creating these layers and layers and layers of the people around them. So you know what happens, what ends up happening, they actually spent 75% of their time fixing things while we spend 75% of time creating things, building things. Because for them the feedback loop is so long by the the time they get there is something wrong they throw more bodies.

Nadiem: That's really interesting. Like so if a culture is nothing more than the collective behavior of expression of the people within that culture, then you can theoretically hire your way out of a culture.

Ajey: Yes.

Sidu: Yes. Absolutely.

Nadiem: You can, if you decided to double your head count within like a year or less than a year. You could essentially remove or destroy whatever vestige of your culture that you had that was working for you, right? The scalability of a culture. Then you reset again from zero. You have to redefine.

Sidu: Then you have noise.

Nadiem: That's really interesting, right? Because I think a lot of people's mindset is about talent. When we talk about talent, we're looking at one end of the equation only. We're looking at how do we hire, hire, hire the best talent, when actually the mechanism for high performing teams in general is also what is fundamentally at the core of their culture. And if you hire too quickly or if you hire leaders too aggressively, then you are diluting that culture by default.

Ajey: I have one more this to say while we're on hire. So what happens is when people start,

Nadiem: or if you're lucky enough to find people who match,

Ajey: Correct, so what happens is when you say hire, hire, hire. If you're not careful about this, humans actually optimize for this metrics, you are measuring them for.

Nadiem: Sorry, say that again?

Ajey: Humans optimize for the metrics you're measuring them for.

Nadiem: Humans optimize for the metrics you're measuring them for.

Ajey: Got It? So what happens down the line, third level when it reaches to a recruiter. His metrics is I have to hire people. So he's focused on numbers. Now once you start focusing on number, their optimize is also focused on numbers, then they're going to hire the numbers. What we should optimize for is skills, right? But that's why at GOJEK we don't hire that many people because we are very particular who we hire. In terms of every year, like for example, when we hire a leader we make sure that they interact with 20 people, not because we are really very anal about it, but because we want to make sure that as much as we are sure that this is the right person for us. They should be sure that this is the right company for them, right? And once I start optimizing for metrics, which is a skill, then you would get this kind of high performing culture and high performing team. But if you start optimizing for the numbers, then you will get high performance numbers, that's all. But that, I don't know how much time you will waste in terms of onboarding those people and how much time will you will waste in terms of figuring out what the hell are they doing.

Sidu: There's a very interesting paradox in this which is, something that I've observed. So, you know, before I go into GOJEK, I've consulted across many different companies, many different skills and something that it took me a long time to grasp was that, you know, we talked about this at the start, that culture are the processes and policies that you do not define. It's everything else because your processes and policies are always a tiny subset of what you actually use for decision making. So when you start hiring very, very quickly, and this is my engineering bias, right? Like I tend to look at this as a distributed computing problem and I look at culture as the software that runs on your people versus the software that runs on your hardware. And when you're running software on hardware, one of the problems you face in distributed computing is how do I synchronize and how do I build consensus? Because I have all of these different servers which are running software with different states. And if they're not in sync, the end result is different. Customers may see completely different experiences because the numbers are different. So in software you have this problem in distributed computing, around building consensus. When you start bringing more and more people into the system, what ends up happening is people take time to absorb the software that's running on other people. Building consensus involves loading that software on to people. You call this onboarding, you can call this induction call it what have you, right?

Ajey: Yeah, you're making it sound matrix man. The movie.

Sidu: I'm a developer. I like to think I'm right. It's my bias. So, so what ends up happening is as you start hiring a lot of people in very quickly, there isn't time for that consensus to build and so what happens is your organization starts devolving to depend entirely on your policies and the metrics, the performance metrics that you set in say your compensation policy is an example of that. And to Ajey's point, what ends up happening is people start optimizing for your formally defined processes and policies, which always especially in fostering companies, will skew the system in very, very nasty, unpredictable ways because the damping effect of culture, which is all of your undefined processes and policies, which is the vast majority of them, does not have time to synchronize across all these new people that have come in. And so the organization starts to skew in very strange, unpredictable ways because you have the small set of processes and policies now shaping everything.

Nadiem: And I think one of the biggest downsides of companies that either don't have a strong or thick culture, uh, or actively foster it which we'll get into in the second half of this discussion. But for those companies, I find that the natural instinct is just to motivate engagement and performance through financial incentives. And that leads to all kinds of problems.

Sidu: Soviet Union, that's what happened to the Soviet Union, centrally planned economies don't scale

Nadiem: I think that gunning for metrics is probably one of the, and linking compensations metrics is one of the most dangerous things to do in management. I'm not saying it's not always a good idea, but I am saying that most of the time that I've seen the bad outweighs the good.

Sidu: You have unintended consequences

Nadiem: A huge amount of unintended consequences. And I think that is the compensating factor when you don't have strong cultural dynamics that foster engagement. And that's the point that I want to say right now. Like strong cultures, I think personally, and please challenge me if you disagree, but I think that strong cultures build the most powerful sense of belonging in groups and teams.

Ajey: They do.

Nadiem: When I can see my behavior reciprocated or mirrored in my day to day interactions with other people, I feel that we share something. When we share something, I feel intimacy and I feel intimacy. I feel like I belonged and when I feel like I belonged. That to me is one of the most powerful, uh, characteristics of a high performing culture. Because, uh, like I read this recent book by Daniel Coyle, I believe, called the, uh, the culture code a really great book. I would recommend it to anyone. And basically one of the concepts that came about there was how high performing cultures stemmed from the sense of psychological safety. Psychological Safety, uh, vulnerability is a big point. There is three things; and then a sense of common purpose. So it's kind of like this concept of belonging somewhere. It's one of the most powerful things to get individuals and teams to behave autonomously. Right?

Sidu: I feel safe making decisions in this environment even though those decisions aren't codified and protected by policy.

Nadiem: Yes.

Ajey: I'll give you an example like last year, Nadiem was debating something an engineering team activity and went up to him and say that is not right.

Nadiem: 'Eng' team is engineering teams, right? People, the audience doesn't know what 'eng' team is.

Ajey: We at GOJEK have product engineering team which is actually, product engineering plus design plus everything, we make them one, right? Because we don't want silos. That's part of our culture as well. So product engineering team went to Nadiem saying this is not right. We're not going to do that. And they put very objectively things. And for the period of time we discuss this like for like maybe one day and then we agreed to yes or what are you guys saying is right? The only way it is possible if the how all these three things which he said, psychological safety, being collected together and we don't have like no sense of insecurity. Those kinds of things could define us that, right? But when you think back now, then you connect; oh this works, this is what articulation is, but if you go overboard or you go little farther and figure out what the hell our culture is, so you'll always have these things you can think and codify by the time you start codifying, you could codify like only 50% of it not the rest of it.

Nadiem: Do you remember when, you know, in the first year when systems went AWOL all the time because we could not handle the load in our first year when the APP was launched and I used to do these massive WhatsApp rants. Where I was deeply punitive in both my language and in my, you know, what was that running joke...

Ajey: "Nadiem is typing ... "

Nadiem: "dot dot dot, Nadiem is typing". Yeah.

Ajey: So if the audience doesn't know if you're in a WhatsApp group, if you have a WhatsApp group and somebody is typing, it comes on the title, saying this guy's typing. And when we had downtime, the only thing to come on top continuously is "Nadiem is typing"

Sidu: Like multi-page rants

Nadiem: Sorry guys, I was a young founder. I was inexperience. Well Basically. I was so verbally abusive whenever these things with the hope of creating the sense of gravity in how bad...

Ajey: Now you know how much indefinite patience I have.

Nadiem: Right. And I noticed one of the things that by doing this, by penalizing mistakes, essentially what I was doing and creating huge amounts of anxiety and frustration, I began noticing that people were sharing less of the mistakes on my group. They were sharing it with other groups, but not on my WhatsApp group. Right. And I began noticing that people were not being upfront with me anymore, uh, in many ways. Like they were kind of sugar coating some of the situation, the gravity of, of the problems. And I began to notice that, you know, uh, they were coming up to other people around me to get advice and help instead of going directly to me. And so in that very early phase of our, of my founding journey, I, I very quickly realized that creating a sense of psychological unsafety was detrimental. Actually, not only did it make people burn out, it also made people lose trust in being able to be open in a form. And because of that stress, I cause other groups started popping up whereby there was psychological safety.

Sidu: You also created managing Nadiem experts. Like there were these people in the org whose expertise was managing Nadiem and so everyone would go through them.

Ajey: They're a bunch of people who actually played that part in last year three and a half years.

Nadiem: Yeah and that's not good because that's extra friction, right? You're not creating a culture whereby people are safe to make mistakes, etc. And you know, I reluctantly change my ways and it really, I saw firsthand the impact on, on culture of sharing downtime, sharing, um, root cause, RCAs and root cause analysis on what happened there. Transparent. And I did something very simple whereby every time, you know, a lot of times over email or when someone, um, said we had an outage or there was this bad deploy or we made a mistake on an assumption of something and we messed up, you know, just simply saying, thank you for this. Thank you for fixing it. Thank you for telling me. Thank you for addressing it. It had a massive, I think I can feel that I couldn't describe how I feel, I just felt this change in behavior that allowed, okay, I'm making a statement as a leader that it is okay and is appreciated because we get to learn all from your mistake. So thank you. Thank you for sharing that with us. It's okay. They'll keep taking risks guys.

Sidu: And one of the challenges I faced as someone who's come from a tech, developer, engineering background who spent most of his time doing non engineering stuff because I work on people, I work on that kind of stuff was a interestingly bringing the RCA culture there. You know, it took, I literally had to become this caricature. I had to become the RCA guy and established this precedent that I will demand an RCA for every failure of employee experience and there will be no negative consequence to that RCA. Right. They will only be a negative consequence to a trend of RCAs where the same problem repeats itself over and over again without redressal. And trying to move that, migrate that culture from engineering to non engineering was a massive endeavor. Like I said, I literally had to create this caricature personality, who's response will always be this, I need an RCA. So I was the RCA guy.

Nadiem: Can you imagine and like business 1.0 people would, companies would actually incentivize that team for how many of those failures got reported and trying to bring it down. So what happened? They didn't get reported. If that's, that's the dangerous part about it. So what you're saying is that you did that and you made it explicit to your team. There will be no consequences to the actual issues that were faced only if it happened over a long period of time.

Sidu: Yep. And my incentives were set on employee experience, not on reducing the number of failures because that just leads to things being hidden because a lower reported number means we're doing well. Whereas positively incentivizing and actually even repeating this idea that what we care about is how do our colleagues feel when they come into the office. How do they feel when they're flying to another office and how do they feel when we have to go to this hotel in this city they've never been to and how do they feel when they come into work led to a much more positive response. And that was purely cultural. There's no process of policy that talks about this.

Nadiem: Right? What, what in your mind is the mechanism by which psychological safety creates higher performance? Why is that? I mean a lot of people, you know, in traditional business, etc, there's always this sense of even a lot of founders, which I and I was part of this group for a while, believe that a certain dose of fear of leadership, a certain dose of I guess bravado and explosiveness, uh, should always, there should always be that slight fear to show leadership. But the more I read books about this, the more I look at behavioral psychology, the more I learned about child psychology and now that I'm a father now I'm beginning to read a lot into that as well, is that actually the fundamental prerequisite of creativity and true state of flow is pure psychological state?

Sidu: Absolutely. I mean, this, you know, this, this opening of mine is somewhat controversial, but I'll say it anyway, right? In systems and organizations where you're dealing with a completely solved problem where the system is clear, there is no evolution and there is little or no need for creativity, fear, skills. And if you look at what you were calling business 1.0, most companies either already had a solved system or they thought they did and that meant that fear actually scaled or they believed it did.

Ajey: I have one more thing on that. So what happens is if you look at the kind of companies there are, these kind of companies did not exist 20, 30 years back? What existed 20, 30 years back for 300 years or 500 years was task based deliver force. Where the workers were optimize for the output and the doctored output did not have creativity at all.

Sidu: In fact, the objective was to remove as much creativity for the from the system as possible.

Ajey: It is instill into manufacturing where you have this conveyer belt and people just do one task the whole day. The sole one board all the time. Right now if you instill fear there, they'll work faster because there's no creativity. Now you take the act practice and bringing it to practice of where you are, how to create something right now when you want to create something and instill fear, what happens? Creativity will have can short cuts, right. And once creativity have shortcuts. You will make mistakes if you make mistake. Yeah, exactly.

Nadiem: And it's not just the creativity. I think sometimes we paint the picture of tech as in it's all about creativity and everyone's in like creative rooms designing amazing things. It's when we talk about creativity to the audience, I just want to be clear that it's not just ideation. Creativity is actually figuring out how to flexibly solve problems that you did not expect.

Ajey: Yes. Multiple Solution for a strategic soft problem.

Nadiem: And in tech, what we call the creative economy in tech is actually because of the complexity, the lack of predictability about what problems we face every day. Forces us to encourage creativity because otherwise we'll end up having to solve it.

Sidu: And in business model although the stuff that you scale by removing creativity and making it into a consistent repeated task and scale that by adding people to the system is now taken care of by hardware. That's the point. Those repeated task we run on computers.

Nadiem: Yes. When the humans doing the value add complex thinking

Ajey: and I did not mean engineer when I said that because if you look at all departments of our company today, so looking at GOJEK as hyper group of startup, all departments of companies today have creativity. Whether it's engineer, marketing, operations. Everyone like even in customer care, they have creativity there because they need to evolve and innovate on how do they serve customers better, how do they automate regular task. How do you actually bring the strategic solution for longterm problems and how do you reduce the issues? Right? Though in companies like us, which did not exist 20, 30 years back, the creativity is a core of the DNA. It does that the bottom or the hard core of the center of the whole thing. And everybody has to perculate that culture. Right? And when you look at creativity, there is no place for anger and ego because if you bring these two things in, it just dies. So that's what happened.

Nadiem: Because it is a very specific subset of creativity, which is group based creativity. Right? It is not, it is not an individual, uh, you know, uh,

Ajey: it's creativity by collaboration

Sidu: I like to say, you know, this is purely my thing that, you know, there's this trifecta that allows you to drive things, which is love, respect and fear. And when you're on the more creative side of things, you really need to dial down the fear and dial up the love and respect.

Nadiem: Hmm. Yes.

Sidu: And if you dial up the fear, you automatically dial the others down and the creativity dies. Creativity is fundamentally an iterative process of repeated failure until you hit success. And if you have fear, then you will not repeatedly fail until you succeed. Because you're afraid. There's a price to pay. Emotional, financial, whatever, but there's a price to pay for failure and that cost is real.

Nadiem: What about vulnerability? What about the thing that is also in the book that the Culture Code Book that I read, there was this whole point and it kind of codified a lot of the things we knew intuitively. Right. Uh, why we instinctively say publicly that, uh, as leaders say that; "hey guys, I don't really know about this. But," or when we go and we say, you know, a "guy, my bad, that was total mistake. I assume that wrong. No, I thought that was going to work and it didn't". And seeing that behavior displayed of saying, you know, I messed up. Uh, and, and that equally, okay, wait, where does that fit in? Where does vulnerability actually fit in into, well, how does that translate into high performing?

Sidu: So for me, vulnerability is a way of saying I iterate to. If I'm, I'm literally telling you that I fail also, you know, it's very dangerous, and it's fine because in this kind of an environment, there are two extremely bad outcomes to not exposing that true vulnerability. Right. One is you risk setting yourself up as this distant unreachable figure who never screws up and somebody who never screws up, will harshly judge someone who does. That's the perception. Yes. Right. So if you hide your failures from your colleagues, your peers, your seniors or juniors, it means that the perception bills that you don't fail, which is obviously false because in our kind of environment where we just do not know the real solutions to a problem at any level of the organisation, you become someone who's unapproachable.

Nadiem: I don't feel like I belong in your perfect little world.

Ajey: I have another way to say this, I always say, and we've discussed this Nadiem as well. I said like expression of vulnerability is manifestation of collaboration. I always say that like,

Nadiem: Say that again.

Ajey: Expression of vulnerability is manifestation of collaboration.

Nadiem: I like that.

Ajey: Because every time you express that you're vulnerable to something you will have your players to collaborate with you to get you out of that situation. If they know that Ajey is weak at this area, I have people who will actually look out for me. Yeah. And if people don't express their vulnerabilities, then they'll fail so badly because the ones covering you there and we have, we have had leaders who actually did not express vulnerability. They actually try to overcome that vulnerability by painting a false facade. Yup. And once the facade goes down they're exposed so badly that nobody accepts that.

Sidu: So you literally took the second, second big risk out of my mouth. Right. Which is a leader or decision maker leaders actually a bad, bad phrase. Any decision maker who hides their fears, will eventually, no matter what they tried to do, end up exposing those failures and then we painted a hypocrite. Then they will just fail like then nobody respects them to Ajey's point and then you're done. There's nothing you can do beyond that point. You, you, you've got a tactical trajectory of success but a strategic trajectory of failure.

Nadiem: Does it have something to do with this kind of sense of camaraderie as well, that if your leader does not share their failures. That means they're not in the trenches with you in the beginning, right? That means you're not taking any, any uh...

Sidu: I mean you cannot possibly be in the real world trenches without failing

Nadiem: You're not in the ring with me.

Sidu: Yeah. You're in an imaginary insulated world where you don't fail.

Nadiem: And that prevents collaboration.

Sidu: People feel when they look at someone or a decision maker who never fails after a while, especially people who worked and been around for a bit start to suspect, this person, this person has fall guys.

Nadiem: Fall guys? As in guys that they blame?

Sidu: Yeah. Because if they've never fail that's impossible. That means every time they fail someone else has taken the the blame. So I probably shouldn't be around this person because the question is, when do I become the fall guy?

Nadiem: And that plays with the psychological safety.

Ajey: I'll give you an example. An early days example. How many times we had down times and how many times after those down times we've had ops chats on the phone for hours and what did we do over there? We collaborated to figure out how do I come out of that, right? Or you had these calls with Sidu or Niranjan or anybody. Or we had group call saying, hey guys, let's problem solve this. If I would've told you no, no, no, no, it's a hardware fault network went down, not my fault as a few times what would have happened?

Nadiem: Lose trust.

Ajey: Correct. Not only that, you would have never ever to bring your perspective as a business person because when you look at a problem, so what happens in, when we look at problems at engineering uh, we kind of have this empathy bias. Like we know that, oh, this developer did not do this thing right. And I know the problem was, I know the problem. I can't problem solve it. So I suppose it would give me a deadline. I knew that this guy can't do it, so I will actually not be able to do it. But once you bring it outside perspective, to me, when there's downtime happen, what did you do? You actually asked me so many questions, you can't call it as a stupid question, but do you ask them very basic questions every time we collaborated on those basic questions and what came out was the solution, which is a very valid solution. It was just a short cut? It came out ...

Nadiem: Despite me knowing nothing about engineering, right?

Ajey: Correct. If I were to not express my vulnerability to you. We would have never collaborated.

Nadiem: And that makes me feel respected because even as a non engineer, you had enough respect to ask me or to follow through my train of thought. And we came up with a solution together, which increases my level of confidence and increases my sense of belonging in GOJEK engineering. That I'm not just an alien outsider, trying to tell you guys what to do. I belong because I helped solve that problem together with engineering.

Sidu: My failures consistently. It took me a long time to recognize that was without intending to I realized after the fact that I was being condescending to nontechnical people, that I was discussing issues with and what I was doing in the process by being condescending was without exposing my, well, I was literally taking a position of superiority, however, unintentionally, which meant that they don't feel respected. Then the collaboration dies.

Nadiem: Yeah. That's the the appreciation point. It's so interlinked with vulnerability. I feel like they're kind of flip sides of the same coin in many ways. Right. To appreciating people. It has to do with the authenticity and the intimacy brought by the sense of shared kind of pain together. Then you cannot replace in any other way. Then there's the kind of this final component, which is the purpose or the mission. Right. Then there's that as well. That is as important as these, these other elements as well. What is it about culture that that resonates so strongly with having a strong purpose or mission together? How does mission affect culture in a team? Does it even affect?

Sidu: I mean, I was skeptical of this, but you know, mutual friends of ours who were founders, uh, you know, who's seen this before many years ago. Once told me; hey dude, there's this developer I know who's just moved to, this was in the bay area when I was visiting. We just moved to the bay area. He's really good. You should hire him. And he was like, he knows me. He knows what we're doing. If he believes in our product, he will call me. But if he doesn't believe in our product, I'm not going to hire him because then there's no point. And over time I came to realize that if culture is your uncodified decision making process, your mission is what provides this unwavering backbone to those uncodified decision making processes.

Nadiem: The why.

Sidu: It's the why.

Nadiem: And I guess, you know for all the times that we stumbled through the initial phases of trying to create a shared culture, what really pulled us all together despite the fact that we didn't focus on in the beginning was that mission, no purpose. People stayed for that, people bled for that.

Ajey: Last week, I was interviewing with somebody. actually no, somebody I was interviewing...

Nadiem: that would be awkward.

Ajey: So, so one of the things they said is like very nice thing they said; like Ajey, I heard all this stuff about your mission and all this stuff. Tell me what are you doing for the environment and people. And that actually got me started and we spoke about like 45 minutes. Like, what our mission is, what we are doing for people, what you're doing for environment. How are we thinking about impacting Indonesians and impacting other countries where we are going and how positively we changed the life of people. And I showed him everything whatever we could do. I could have not done that if I did not know the mission and they would have not got influenced by us if they did not know that we really truly believe in what we are doing. One thing is that and second we are acting on it. And that thing of once you have that thing mission, our mission, what our mission is, our mission is to go and do the right thing for the drivers, impact their lives. That means means you will never do something wrong for drivers. That becomes our culture.

Sidu: I mean we actually have this as a tagline on our employer branding in some of our campaigns. Nice people, hard problems.

Nadiem: Nice people, hard problems.

Ajey: Like for example, if we are talking about GO FOOD right and we are looking at GO FOOD and how can we be more environment friendly within GO FOOD. Like GO CAFES is a good example. GO CAFE is very good cafe we have by the way you should go and see it. So GO CAFE. We transformed the GO CAFE from being not very environment friendly to more friendly now. We have every biodegradable product over there, we don't use plastic. If you go to our offices, we don't use plastic right? And that actually has a very big mission change.

Sidu: We never set this as a instruction. I remember in India office, at no point have I ever mandated this, but one fine day, all the disposable cups went away. And there were these reusable steel mugs that showed up beautifully designed. And I had nothing to do with it.

Nadiem: Yeah, it wasn't a command, it wasn't a policy. It becomes emergent behaviour. And I think that's the most salient point about culture. It is the most scalable and viral way of generating either positive or negative behaviours. And it creates that powerful sense of shared identity, which is just another word of saying, I belong here.

Ajey: And that's why I said culture is the expression of a group.

Sidu: Yup. It's why I always anchor on on, on things where I say I'm most proud of stuff that I had nothing to do with. That's, that's the stuff I'm most proud of.

Nadiem: That's when you know culture's kicking in. And alive and well. Guys, we've run out of time, but we could go on hours on this subject. Thank you so much for the insights. Please come again on the GO FIGURE podcast.

Ajey: Thank you for having us and we'll have more fights on this.

Nadiem: Awesome. Awesome. Thanks guys.

Ajey: Thank you. Bye Bye.

Outro: Hey guys, hope you enjoy the podcast. If you liked it, please hit like subscribe and follow us on social media. Thanks so much for tuning in.

See Full Video