Home
Contribute Contact Us Browse all meetings
Home Contribute Contact Us Browse all meetings
GitLab / Remote Work / 26 Sep 2019
GitLab / Remote Work / 26 Sep 2019
Previous Meeting Next Meeting
⏯
Add meeting Rate page Subscribe
youtube image
►
From YouTube: Remote work benefits with Prof. Raj Choudhury (Harvard Business School) & Sid Sijbrandij (GitLab)

Description

Prof. Raj Choudhury talks with GitLab co-founder and CEO Sid Sijbrandij on the topic of remote work.

Related links discussed:

Harvard Business School - How Companies Benefit When Employees Work Remotely: https://hbswk.hbs.edu/item/how-companies-benefit-when-employees-work-remotely

GitLab's values: https://about.gitlab.com/handbook/values/

GitLab's main all-remote culture page: https://about.gitlab.com/company/culture/all-remote/

Being intentional about informal communication (social calls, coffee chats, etc.): https://about.gitlab.com/company/culture/all-remote/informal-communication/

Working asynchronously: https://about.gitlab.com/company/culture/all-remote/management/#asynchronous

Scaling by documenting: https://about.gitlab.com/company/culture/all-remote/management/#scaling-by-documenting

Advantages and benefits to all-remote: https://about.gitlab.com/company/culture/all-remote/benefits/

Drawbacks to all-remote: https://about.gitlab.com/company/culture/all-remote/drawbacks/

Hiring in an all-remote organization: https://about.gitlab.com/company/culture/all-remote/hiring/

People who are adopting a remote lifestyle (speaks to caregivers, military spouses, travelers, etc.): https://about.gitlab.com/company/culture/all-remote/people/

Learning, development, and onboarding in an all-remote environment: https://about.gitlab.com/company/culture/all-remote/learning-and-development/

GitLab's handbook first approach: https://about.gitlab.com/company/culture/all-remote/learning-and-development/#handbook-first

Advantages, challenges, and the primary differences between all-remote and other forms of remote working: https://about.gitlab.com/company/culture/all-remote/part-remote/

A brief company history: https://about.gitlab.com/company/history/

GitLab team page: https://about.gitlab.com/company/team/

GitLab organizational chart: https://about.gitlab.com/company/team/org-chart/

GitLab product stages, groups, and categories: https://about.gitlab.com/handbook/product/categories/

GitLab Contribute (every 9-12 months, we gather all employees in a single place for a week of bonding and informal communication): https://about.gitlab.com/events/gitlab-contribute/

GitLab Visiting Grant (a travel stipend for team members visiting other team members): https://about.gitlab.com/handbook/incentives/#visiting-grant

A

I am Sid, co, founder and CEO at get lamp, and today I'm, together with Raj who's, a professor at Harvard, Business School and he's written a great article about remote works. So we're gonna talk about that article and his insights today. He might ask me some questions and right before we proceed to you, maybe introduce yourself and give a bit of background sure.

B

Thank you said so. My name is Raj Chaudhary I'm an associate professor at the Harvard Business School, and my research really looks at geography, mobility, but also in mobility. You know the reasons people do not move when their economic incentives to move and in that context, what I?

B

What I've been really interested in the in the in the recent past, is to study this new form of remote work, which I call work from anywhere which goes beyond the traditional working from home model in allowing people to choose the geographic location they want to live in, and what I argue in my academic paper is that this form of remote work not only gives you the the flexibility with respect to how you spend your time, but it also gives you the flexibility to choose where you live in and so, for instance, you could move to a lower cost of living location.

B

You could move back to where your family is from. It could help resolve dual-career issues and so on so forth. So this is the agenda that I I really want to study going ahead. I of course understand. This may not work in every organization or every kind of task, and that's why I'm, having all these conversations with with companies and managers and CEOs just to educate myself as well in terms of what are the business model, innovation is going on around good practice. Also.

A

Thanks so much and thanks for joining us and being is broadcast, I am with you. I fully believe this is like the future. We call it all remote and it's like the company doesn't have some offices, it's not that you work remote. Some of the time know everyone is remote and your remote all the time.

A

What is driving your interest in in researching the impact and expansion of remote work? Why did you get into in this? That's.

B

A great question said so: my personal intellectual journey has been suprotec. Academia was a consultant with McKinsey for many years.

B

I was remote all the time and then I was running a business for Microsoft in Asia, where I was traveling all the time back to Redmond, and so when I got into academia really I started researching geography, mobility, because I saw the immense benefits that moving people around had for multinational firms, but then in the middle of that exploration, I also realized that people do not move all that often, and there are many reasons why they do not move and academically I call that framework a framework of mobility frictions.

B

So they could be regulatory reasons are related to these sirs. They could be family reasons. They could be personal reasons. I want to be close to my hometown or aging parents. They could be karmic reasons that I don't want to move to Silicon Valley because it's so like expensive to get a place to live there. So if you believe the two things that I believe that moving people around is productivity enhancing, but there are frictions at all levels in moving people around I guess the question then becomes.

B

How do you solve this problem of frictions and I see working from anywhere as one of the possible solutions? So if you cannot hire an engineer because that person would not get an h-1b visa, could you put that person in Vancouver and make that person work into Silicon Valley remotely? So that is how my interest in this topic progressed and what I'm now exploring said is: when does this work? What kind of jobs does it work for and what are the boundary conditions?

B

So when does it not work and at the I want to say like at the beginnings of my exploration, so I'm I'm also here to learn from you guys and from like this discussion.

A

Cool and I can't help, but giving my quick take on that I, don't think it there's a benefit in moving people around per se. I do think, there's a benefit of interaction across many a diversity of background. You don't necessarily have to move yourself. I think where you live should be it's very.

A

It has a big impact on your family if you move like ten times during your childhood, that that is, that is tend to be not a great thing and there's like family to care for there's aging parents who take care of there's people who just want to live somewhere because they like they want to explore that a lot of people and give I have that too, and, of course, their spouses who are bound to a certain place.

A

So if you deal with something in the physical world- and it can be anything from a hospital to a hardware manufacturing, it doesn't make sense to be all remote. It makes sense to gather around those atoms, but more and more companies are don't need any physical manifestation like everything their inputs and their outputs are completely digitized and I. Think for all those companies it could work. I do think you have to accommodate remote. It's not like just like when you're co-located, you take care of an office, and you do things like that.

A

If you're all remote, you have to like be intentional about the informal communication and that happens in the company, so.

B

Just building on that so I'm completely with you and so I'm a big advocate for this. So don't get me wrong so and the context that we studied in that academic paper and that I wrote about in the Harvard Business Review was the United States Patent Office. The reason they thought about this work from anywhere kind of policy in that organization was it's very expensive to get real estate in Alexandria Virginia, so they were motivated by real estate savings.

B

But what we did is we conducted this really detailed examination of what happens when people move from Alexandria to Lake, now Florida or Texas or anywhere in the country, and you know even that you're always concerned about whether the productivity effect you find is a correlational result or is it a causal result?

B

We exploit what we call this natural experiment that helps us identify causal results and I can talk more about the experiment if you're interested, but we find that productivity goes up 4.4% when you move from a traditional work from home program where you had to still come into Alexandria one day a week from that setting to this work from anywhere setting and what we argue in the paper is that this really works for them because of two reasons, because first, the job of the patent examiner is a very independent job.

B

You don't you're, not working in a large team or just examining a single patent over like two or three years, and the second thing that worked for them was people had to first spend two years in the office before they went to the work from home program, and then they could move to this work from anyway program. And so what that does is it takes care of the fact that you have a social network in the office? You know who to call when you have a problem.

B

So here's a question for you, since so in a setting where you are working in a team or in a setting where you're bringing on someone new to the company, have you thought about how you resolve the coordination costs and the learning cost? So they call them if you are always remote from the beginning, yeah.

A

I think so to solve for kind of learning, there's a lot of things. You learn implicitly at an office that we have to be explicit about, and that's why we have a company handbook of over 3,000 pages, where we describe every process and also like who works on what, like a really big part of. It, is like being able to find the right person, so we have a team. Page is even public.

A

Everyone's listed, we have a few Google gitlab categories, you'll find a very extensive page of all the different parts of our product, who's, the product manager for that who's, the UX person for that etc. So we do a really good job of detailing, or people are experts at. So you know who to reach out to so. You don't need to be a the office to find that out and that's it by the way. That's a super inefficient method because, like it's not that people are at the office and they naturally discover this.

A

No, they have to tell people on the shoulder like who works on this. Who works on that, and our policy is just write. It down write it down. People have a nickname, great write it down. You don't have to be at the office to find out and then regarding independence, obviously like your most, if you're completely can work independently that works well, but I think at gate lab we've, we managed to do very collaborative tasks well, remote and to values that get lab help with that.

A

One of them is transparency, and so we all work in the same issue tracker every everyone is visit. Everything is visible to everyone and the second thing is iteration. Iteration means scoping down tasks, so you can complete them faster and doing many iterations quickly and by scoping them down. You need to coordinate with fewer people and when we work together, for example, front-end and I can teach I'm completing something they do have a video call together to kind of split up the work for the next two weeks, but they don't.

A

They don't have a project manager kind of doing that for them there's a huge degree of individual responsibility, and you mentioned real estate savings I think the savings are not just the office costs, but an ability to hire from a much larger pool the productivity that you have when you. You cannot be shoulder tap like when you, when you're are in control of your working environment and and how you plan your time.

A

The happiness imbalance that comes from not saving time on the commute, commute time and and divorce rate are correlated and the ability to kind of not have to churn out of the company. When you you need to move someplace else or your spouse, or something like that. So.

B

A.

A

Retention so.

B

Just add to that, let me just add a couple of interesting findings from the setting that I stress, so we found that people average moved to a lower cost of living location and we estimated that the US Patent Office did not reduce the salary, because you were signing up for this program, so the net income increase was actually 3%, which is not insignificant.

B

So.

A

This is real money in.

B

Your pocket, the second thing we found is that there is they're very interesting patterns just in one second off, where people move to so just to give you one example: we found this cluster of people who move from Alexandria Virginia to Florida, and we, when we looked at the characteristics of these people, we found that they were typically older, more tenured people and I I've thought about this result and I like it a lot because I think, if you think about the aging workforce problem, we have, if working from anywhere, could extend the productive life of our workers by even a few years, and they are as productive or even more productive.

B

If they move to Florida or retirement locations, then that could be a really really important way to sort of boost the productivity not only for companies but for the economy as a whole on the flip side, so once again, I'm a huge proponent of this policy, but I'm also thinking about all the possible boundary conditions. So when I talk to today morning, I was talking to a CEO who was run this really large electronics company, and they, you know, gentlemen, said Roger.

B

Maybe it's I'm old, but you know my bias is this model will fail for one reason and that's what a cooler conversations when you are in the same location you are, and especially when you are CX or innovation driven person. You get these spontaneous conversations in the parking lot or in the water cooler or you know, at the end of a meeting or the beginning of a meeting, and that had me thinking whether an all remote model can find some alternative solutions to stimulate these spontaneous conversations. Yeah.

A

I appeal to that earlier, like people take a lot of time there in her office and everything like the rent, a parking lot. Everything just put a tenth of that energy in facilitating informal communications. I'd get my way of super intentional about that. Every day we have a group conversation about a certain aspect of the company. There's a big slide deck. They don't present anything just Q&A from everyone. Lots of great ideas follow from that.

A

Every day we have a breakout call 20 minutes with a group of team members from all across the company kind of a fixed group of people, and today their question we were asked was: what's your favorite purchase you made in the last year and saw some great paint-by-numbers paintings by some team members, not all of them finished and I, actually gifted someone a a USB coffee, mug anyway, just informal communication. We also do coffee chats where you plan twenty-five minutes with a core team member and just to hang out together.

A

No agenda can be talking about. Work can be about private life, but it's not weird to put it on someone's agenda and we people who join the company. We have them do that a few times, and the last thing we do is group socials.

A

So every week you hang out for 50 minute with the rest of your team. You just there's no agenda, you mostly talk about life outside of work, and some people have a beverage as well. Well, they do that. So there's lots of ways to do this and making people commute for 90 minutes. In addition to the time it takes to do the water cooler talk is one of the most inefficient ways to do this and it's ridiculous that people feel well. It's not ridiculous.

A

People feel uncomfortable kind of organizing this, but, as we know like, if you organize something, you tend to get better results and why depend on randomness for that I know at Pixar they place the restroom centrally, so people would bump into each other. Yes, okay. These are. These are good ideas, but why not step it up a notch and actually organize the informal communication? And we do this? We have a lot of things we do also to have people meet in real life, but I think it's.

A

It can be stimulated it's working out and- and you still say for the commute time, because look you want informal communications, you don't want people sitting in the car by themselves. Sure and.

B

Just to add a couple of thoughts with that I think.

A

That.

C

That's.

B

An excellent point and I would love to talk more to you about this offline. So the other thing that I thought about is so two more thoughts to add. You know I've found in the US Patent Office context that people were forming these informal communities of practice. So let me explain what I mean by that. So you were no no longer mandated to live in Alexandria Virginia, so you move somewhere.

B

But if you had people from your same technology group what they call the art unit in the Patent Office within the 25 mile radius of where you lived, we actually found an even more disproportionate higher productivity increase, so the average increase was 4% 4.4. But if you had someone from your art unit living within 25 miles away, you are now living you had even a higher productivity degrees, and then we are still digging this, but at least anecdotally.

B

What you know one group of examiner's have told me is that they now have these water cooler conversations, but on a golf course, so they have informally found- and this is completely based on individuals seeking out they've, been informally form. These informal communities and they meet once a week or once in two weeks, play golf together, but also talk about work.

B

The larger question I have in a it's a question for you, so I've done some research with very large multinational companies like Microsoft and one of the things that seems to work in in those contexts where people are spread out across China India Israel Redmond is temporary colocation that you don't need to move people permanently.

B

You go to Redmond from India for three weeks or you create these temporary events where developers on the same technology all come together. They do api compatibility testing, but they are also going out for beer they're getting to know each other much better and when they disperse they still retain that social connection. Have you guys thought about bringing people temporarily together, physically or you think in your model? Even that is not required. Yeah.

A

For sure we do a couple of things, so one thing is we do the local meetups. So if there's multiple gitlab is in your area, it tends to be a local meet up every now and then, once a month, just go to a co-working space together have dinner. Afterwards we have a travel stipend. So if you're visiting other get lap team members around the world, you we partially reimburse you for the travel expense that that cause.

A

We also bring the entire company together for a week every 9 to 12 months, and then we don't do work because it's a super expensive thing to do so. We focus on the informal part. So we what we do is we do excursions together. We tend to go to an interesting place and we have an unconference where we talk about things proposed by individual team members in small groups, kind of informal communication.

A

So we do the things that that being in person is best suited to and I think that works really well and then also we we do throughout the year. We do a lot of like video calling and you can watch our YouTube channel like we do bring people together and I. Think there's something to in person. That is special and we do stimulate that in a multiple ways.

A

But it's we don't want to rely on that for work and we don't need it for work. We do we need it because it's it gives the feeling of belonging and it helps to get to know people. But it's not a necessity.

B

Let me ask you this said so often, multiple to.

A

Any board members watching it is a necessity and that's four and a half million. So thanks for that, we we do invest there, so it's not needed per se, but we do believe in the incremental returns of that.

B

That's very interesting, so just up one one question that I had for you was of all the reasons that I can think of that I have documented in my research the problems with leases, the dual careers, the. What prompted you to first. First of all, come up with this model, so real estate savings- or you know what was the the first reason you thought about. Is this model yeah.

A

I, like most good things that get rapid it kind of was other people's ideas and basically people stopped showing up at work and so I started with the first team member in Serbia and my co-founder in the Ukraine or remote by default. I hired some people in the Netherlands. They came to my place. I had to desk for a couple of days and then they kind of stopped showing up and I never made them, and they just worked from home and after he graduated from Y Combinator.

A

We they told us I should get an office because there's a remote thing only works for engineering, so we got an office and then all the executives are hired. We're like. Oh, it's not going to work for my function, so I need people here. I'm like great. We we bought a lot of desks, uh-huh we're ready and then then yeah everyone kind of stopped showing up after a couple of days just because they were not missing out on any information. They were not missing out on career opportunities or gossip or contacts and I.

A

Don't think people come to the office because they like commuting they come to the office because they don't want to be a second-class citizen. I think that if you make sure they're not missing out, they don't necessarily want to commute to the office and.

B

One thing I've thought about, and you know something you said triggered that is it important that the cxo signal this behavior, because you know if the CXOs are always in one single office, then, if I'm, a middle manager and if I'm not in that office, I might feel I'm missing out on a second.

A

Exactly and we had that dynamic, even with our board meeting like if there's people in person like the restaurants to be there as well, also because hybrid calls for some people are like on a shared video camera or just really horrible. Anyway, we don't have that people sometimes suspect I have some executives in my house so I.

A

We have be me here, a telepresence robot, where everyone can log in drive around my living room and they see that only the CEO shadows to a Danielle and Eric are here, but we don't like executives, don't congregate and that's so important, because if you don't do that, then, as executive you're gonna adapt the remote working practices, you're gonna have things in public channels in chat, you're gonna use, get lamp issues, you're, gonna use, Google, Docs, you're gonna have recorded meetings that you can easily live stream.

A

Every single person needs to adopt those practices, and it's so much easier to do with the whole company and a lot of companies that are hybrid remote, really struggle because, like part of the company, depends on all those in on those well depends on it. They basically don't have the discipline to record them, write down things. Then all the remote workers I feel like they're, missing or they're, not feel like they actually miss out on information. That's essential to do their job. Well,.

C

And so for those viewing I joined, give up earlier this year to curate and lead and amplify the all remote message get loud, but for the past decade I've worked in remote environments of all types, and this is the first time that I've worked in an environment where it's all remote and I'll say pert. From from the employee perspective having the cxos also be remote, it sends an incredibly clear message that you're not going to miss out on potential praise and promotion opportunity because you're not in the office.

C

It just removes that barrier altogether and when I've worked in other situations, where there was at least one headquarters that some of the higher profile executives commuted into you as a remote worker. You definitely wondered if that lack of face time would impact you in some way and so to relieve that mental burr it it massively helps morale as well as productivity and what I've seen that's that's fascinating.

B

And so I, you know I I, guess it's one. You know. One thing to think about here is: it seems, like you, guys, have really subscribed to this model and by the way, I totally agree with what both of you guys said in our research. We actually find empirical evidence of that. So we compared the productivity of people in three stages. The first stage was when they were in office.

B

The second stage was this traditional work from home, where they were four days of the week somewhere else, but there to come back on a Friday trans, Andrea, bergenia and the third way was when they were completely cut off, like they could be anywhere, and we find that the productivity gain was much higher in the final step. So what we argue in the paper is what we call cut the umbilical cord. You know this piecemeal remote thing doesn't work. Oh, it doesn't work as effectively as this complete flexibility, yeah close the office yeah.

B

So let me ask you this sense, so it seems you're really subscribed to this model. Is there as a CEO? Do you still have any sort of strategy question around this like any unresolved question around this model, or do you think it's it's something you have perfect I think.

A

That it requires a lot of discipline and every single day we pay attention that people do it in the way we prescribe. For example, handbook first is a really hard one. People tend to send emails like oh. This is now like this or give a presentation. This is now like this. No, you should change it in the handbook. That's the that's! The way we keep kind of the documentation up to date so that everyone can rely on that for how to how to work together.

A

That's you should do that too in another company, but for us it's essential and we that's a daily struggle. The other thing is it used to be that if I search didn't believe in that now the investors are, on the other side, we're actively looking for more all remote companies, because it's have lower costs and they can hire people. They can hire better people. That's.

B

Right, I totally agree. So so let me ask you this like differently: do you think this is scalable, so I wonder how many employees you have right now and do you think, is this like infinitely scalable or is there like a sort of like inflection point beyond a certain.

A

Size yeah so I think it scales better than the other model. So it's not that we have to defend like oh does. This scale know like I, see that a team in one room works really well. A team on one floor probably works. Ok, team on multiple floors. You start having some problems: team and multiple offices getting harder team and multiple cities getting harder team and multiple continents getting super hard. That model breaks down. There's there's very little benefit to colocation. If you're such a big company.

A

Well, the benefits of remote writing down your process is stimulating cost company informal communication. Those could get much more pronounced at scale. So it's not just that their skills. It skills way better right.

B

And that's very interesting because in my world there was a seminal research by Alan at MIT, which showed how the communication patterns between individuals you broke down. If there was a wall between them, so even the same floor or a staircase between them, so I think the point you're making is really well taken that when, if I'm physically in one location, it's not that there are no coordination costs. So that's really interesting. So I guess you know, I would love to keep this conversation going and talk to you offline.

B

But do you have any questions right now for this live stream, other thoughts or questions yeah.

A

Do you think that all remote can have can help to kind of? So how be part of the solution in cost of living traffic and housing crisis is that we have in the Bay, Area and and spreading opportunity more equally around the globe? I think.

B

I think it's it's definitely something we should take very very seriously, and it's for all of these reasons that you mentioned, but also because of the immigration situation right, you know, I can tell you how many anecdotes of you know. Students and Friends I hear these days, who cannot get a US visa or are worried? What's gonna happen, it breaks, it really happens without a deal.

B

Where should I go and I think for all the reasons that regulation all the uncertainty that regulation creates and all the reasons that people you know have on a personal or a family. You know dimension I think this is. This is something that we should really really take seriously and we should keep as academics I'm going to keep pushing this one one, you know think I've mentioned I'm doing the study of a very large multinational, and what we find is that one of the sort of frictions that is very hard to circumvent is temporal frictions.

B

So, if you're not in the same time zone, it's very hard to at least stimulate some of these even informal or spontaneous communication. So if you're not like it's very hard for people to to sort of wake up in the middle of the night and get them to do informal, so I was thinking, maybe as you scale globally. So one of the insights of that piece of research which I can talk about more if you're interested is, there was much more communication.

B

If you were in the same time zone, which probably implies that Silicon Valley should look more closely to Vancouver rather than to any location, east or west. So it's easier to do these communication patterns north-south than east-west, but I think there are, of course, many wrinkles that will come up along the way. So one thing I have thought about and I'm glad you mentioned.

B

This was: can the CX community start doing this as a signaling device for middle managers, but also when you, potentially, you are able to put your CEO in the location in a different location than the CFO based on actual market needs, so I know go, for instance, put their CEO at one point in Bangalore and my recollection is the CFO was sitting in Europe and the CEO of course continued to stay in the US. So I think this is a really interesting trend in work. From my perspective, we need to understand it better.

B

We need to have more success stories. We need to write about it and you know I'm really glad that you guys are actively living this. This model cool.

A

Yeah and time zones are the bane of our existence. They make everything much harder. So a lot of our values like transparency and iteration. They help to work asynchronously because that's what's needed to overcome time zones. It's not waking up in the middle of the night right.

B

And.

A

Darren, you might want to write this up on our pages. I think they're scaling conversation we had earlier is interesting, but I also think that maybe we should talk about the stages of remote working. You have companies with no remote. We have companies for remote is kind of allowed like you can work some days from home if companies that allow some people to work remote all the time they have.

A

Companies that are completely remote but coordinate across to one timezone and envision is a an example of that, where they mostly keep it in Eastern time they have companies that were try to work asynchronously across many many time zones. I think that's a sliding scale and I think what we're seeing a get lab. We're attracting a lot of people that worked at companies where they were allowed to work remote full-time, but where there was a hybrid remote organization and those people tend to be second-class citizens and then flock to an old remote organization.

A

And then asynchronous is the hardest. And we haven't totally nailed that our workers in asia-pacific are not getting as good of an experience as the people in Europe and the Americas and we're continually kind of trying to make that better and and to do more asynchronously. So I think that I will be interesting and I. I hope you.

A

You can use that differentiation in your research between the different types of remote working, because I think there's a big difference between a company like WordPress, where the management doesn't even have an office and an a company like github, where all the senior leadership is is co-located.

A

So.

B

Now I think I think we're thinking about the same question said really happy. We have this conversation and I would love to talk to you offline. So maybe we can set up something and I can come over and you a person. Let's.

A

Let's stay in the scope, just to close the call thanks so much for taking the time thanks for studying this I think it's a huge benefit to the world. Sure.