 Thanks for joining. This is an AMA, stands for Ask Me Anything, and I especially appreciate hard questions on things that are not going well, culture, pay, team composition, things like that. Clement, thanks for starting us off, Clement. How did I get connected with my current mentor? There's a couple of people that I learned from. We have our investors, especially Bruce and Vili, the A and B round investors, because they work at a venture capital firm and they have a board seat. They spend a lot of time with us. Then there's advisors, advisors to the companies, so that's Luke Canes, ex-CEO of Puppet. There's Zach Holman, prolific programmer, and there's a third one, Zach Erlacher, an ex-CEO of Zendesk and Duo Security. But probably the most important one is my CEO coach. His name is John Ham, and he got introduced to me, I think, to Bruce, and he's been a CEO himself. Jim asks, what are you most excited for for the year, either personal or business? The personal thing already happened. We moved to a better place. I'm excited about that. We just had a housewarming party for people living in the Bay Area yesterday evening, so that was fun. But business-wise, I'm really excited about completing our vision of not just making a tool for developers, but also making something for dev and ops. Let me share my screen. If you look here, all the ones in bold are things we're adding to GitLab this year. I think we're on track. We have added about half of these, slightly less, but the year is not half as well. I think it's so... We stumble across a secret. It's a secret not in the sense that we don't tell anyone, but it's a secret in the sense that people won't believe us. The secret is that it's so much better to have a single application for all the steps in the DevOps lifecycle, all the way from managing to monitoring and securing. We didn't believe it. Camille had to first convince Dimitri that it was better to integrate GitLab CI with GitLab, and then they had to convince me, and we both reluctantly agreed, and then it turned out that it's a lot better to have your CI integrated, and we thought we could get all the benefits with just a really great API integration, but it turns out, if it's a single application, you're able to work from one interface, you're able to message back and forth. You don't have to pass credentials around. You don't need another login, or you don't need another session to set up. It's just better, and I think that's our secret. That's what makes GitLab unique, and now being able to do that across more parts of the lifecycle is really amazing. We're trying to kind of write down a couple of like, what are the emergent benefits of a single application? And there's probably a hundred, but there's a lot of things that just become easier, post a link in the chat. So that's what I'm very excited about. Then of course, I'm also excited about where the business is going. Matt is visiting San Francisco from Colorado, and he was at housewarming yesterday, and me and Michael McBride were talking. We're going from a feature cell to now we were talking to a, he visited a potential customer, and he's like, well, yes, they were interested in GitLab almost to save money, but now the discussion is going about, okay, well, how important is security to you? How important is it to speed up your DevOps cycle time? And for them, it takes about six weeks for them to release something. If they can cut that down to two weeks, it would mean a kind of a major improvement to their business. They'd be able to launch much faster, get new ideas out faster, get things tested faster, and they could kind of outmaneuver the competition. So I think us, instead of selling, hey, we're GitLab, we're selling version control, we're now moving to, hey, we're GitLab, we're selling complete DevOps, we're now moving into, hey, we're GitLab, we're selling concurrent DevOps, we're selling you a 200% faster DevOps cycle. And that's interesting, because suddenly you are solving a business problem. Like, what more, like, if you tell executives like, hey, you now have to wait six weeks for something to get done, it could be reduced to two weeks. How would that, what would that do for your business? And most of the time, the answer is a resounding, wow, that would be awesome. We'd be able to generate more revenue that way. So I'm really excited about kind of having that conversation more, because I think we'd be helping customers more and we'd be able to generate more revenue for ourselves too. So Clemente asks about a compensation discussion. Yeah, I haven't followed the whole discussion. I see, I can, I understand what's in the description, I've seen the issue, but I haven't read the whole friend. I think in general, there's all kinds of ways to kind of improve, improve kind of our ability to attract and retain great people. And we have to do what's most efficient. So efficiency is one of our values. So we have to kind of balance the different things. And right now, I think, two things we can do to increase efficiency is to reach out more. I know most of the people that are currently in the company probably opted in to get lab like they, they, they, you found get lab instead of get lab finding you. I think in, in general, and that's, and that's, that's awesome. But I think in general, like for example, when we do an executive search, it's, it's with executive searches, it's like not possible to say, Hey, we're open. We're hiring for this that are that what you do is you go out and you find the best people and you have, you, you approach them. And I think we'll always have a mix of kind of people that really like want to work here and that found us, but you got to recognize that most people are not aware of get lab. So we're missing out on a whole bunch of candidates that don't know about us. So actively going after people, I think it's really interesting. We've got a great story to tell. We're a very interesting company to work for. And I think we should, we should tell that story more often, just like we, we don't just rely on inbound sales. We, we have SDRs that reach out. I think for, for attracting the best people, we should also reach out. We'll have the people like most of you that joined that found us that really are interested and that joined and that should always be a healthy mix. And actually, I think we should also probably work more on employer branding. But I also think we should reach out more. So I think that's, that's a very cost efficient way to do it. Andreas asked, how did we come up with the name get lab? Mitri came up with it. He named it and he created the project a year before I even know it existed. And I think there's a common ancestor in get web. So get the program used to call or still comes with something called get web, which was a quick way to browse repositories. You could set that up on a local server. And you can say that probably get hop took inspiration from get web. And then we took inspiration from, from get web and get hop probably to come up with get lab. David asks, where I want to travel to? Yeah, I think very high on the list are Japan in general, not a specific place. Although I'd love to see Tokyo. And I'd love to go to Japan because it's a very kind of advanced civilization and culture, but very different from, from us. And, for example, the street naming, like we refer to things by, by street and number and they refer to, to locations by the building. Like, it both makes sense. It's just the opposite of what we did. And I think that's really interesting. And the other one would be Moscow. Not sure if I want to go or my, my wife Karen just really wants to go there, especially see the beautiful subway stations there. I think maybe the inverse of the subway stations in San Francisco, I live near Civic Center. If you want to Google Civic Center Bart, then on YouTube, you'll, you'll find out what I mean. Philippe, if you could go back in time and change one thing and get like, what would it be? I'd get more serious about get lab.com sooner. Now, how soon? I'm not sure, because like you have limited resources. So was it 2015, 2016, 2017? I'm not sure. But I do think right now that we shouldn't, we're not where we should be. Now, I'm very proud of the team and we're making great progress. I just was in the GCP migration call and it looks, we made like 25% progress towards the migration. So with very kind of being way understaffed, I'm very proud of the team and I like the progression now. But it's the progression is nice, but where we are is not where we should have been in this part of our life cycle. So from an investment perspective, hiring and people, it was too little too late. Kim says Summit 2019 in Japan. Kim, in Tokyo, I think an Apple cost like 10 bucks or something. Like it's, it's, it's not outrageous, but it's, it's a very expensive country. Joe asks, how much, how many hours do you sleep each night? Try to sleep eight hours, sometimes a bit more, sometimes a bit less. But that's about, we're getting, we're kind of getting that much sleep. We go to bed around 10 o'clock, probably sleeping by, by 1030 and then wake up at 630. And then within 10 minutes, we're at the elevator going down to the gym. I've, I've went to the gym every single day since we moved here. So I'm on a streak. Clement says, I noticed we changed the handbook to set a target date for the IPO. Yes, that's a, that's a, that's an interesting one. I think it's Wednesday, November 16, 2020. The fact that we made it more specific doesn't imply that it's more likely. I think what we're, what we're saying is we're going to strive towards that and we can decide at any point to kind of delay it. And that might be like most companies end up delaying it because it's, there's, there's lots of downsides and upsides to it. If you want to get an idea about that, there's a document called IPO 2020 pros and cons. Oh, let me make sure I message everyone. There you go. And there's the pros and cons. So we're not sure whether we, we do it. However, we're sure that we're going to drive towards it. So we're probably going to start kind of the, the, the, the legal process, but like the preparations for this in the fourth quarter of this year, because you kind of need two years to prep and it's going to be about a million dollars to do all the things. We could start a year later and have a year to prep, but then it gets twice as expensive. So now you need two million dollars. So it's the earlier you take a decision, like we might do it. The, the kind of the more time you have, the easier it is and the more affordable it is and the less stressful. So considering that we're going to, we're striving for it, what would be the earliest time to get out? It would be when you have a million, a hundred million dollars in recognized revenue. So that's not the same as ARR. That is kind of ARR is like your run rate or even the recurring revenue, but you kind of, it's like past revenue that you recognize. So for example, when we make a sell, like we say, Hey, it's an incremental ACV of like, so suppose we sell subscription for a million dollars, it's a million dollars incremental ACV, but we recognize that over the course of the year as like a 12th every month and we need like that recognized part, we need a hundred million of that and pulled it the stats and came like, we could probably go out four quarter of 2020. And yeah, so it's kind of late and then you want to be before Thanksgiving. So this was kind of the date and I've learned with setting a date of 2020, it made all our kind of kind of company and leadership decisions kind of easier. You say like, Oh, 2020 want to be here, you can plan back from that. And I found that planning back is a lot easier than planning forward. Just planning forward, you don't know what whether the curve is going to be like this or this. So then you're debating the whole time. If you know you've got to go here, it's much easier. So planning is much easier if you know where you want to go. So we'll keep it. You now have a specific date in the year and it makes planning for the executive team really easy that any point we can decide, okay, we might be ready to IPO, but we're not going to do it. So that's, that's the current thinking. Oh, and how the date was decided. Yeah. So as late as possible in Q4, but before Thanksgiving, what are my favorite snacks? Wow, that's nice out of the left field. So I drink a lot of Soylent because it's kind of delicious. It's like tastes like chocolate milk, which is amazing. And it's kind of healthy. And I don't have to spend a lot of time prepping food. However, reminds me for lunch now, I currently purchased a couple of gobble soups. So I'm going to try a soup today for the first time in a quarter or something. And gobble is this amazing service. I know there's a thousand meal services that gobbles really, really legit. And I eat better at home than at fancy restaurants. It's amazing. So looking forward to that. And then I don't have any other snacks in the house, because if they are there, I eat them. And we have a problem because there was a housewarming yesterday. So I think there's lots of stuff left. It's kind of healthy. So but I think I'll start nibbling on that. But very frequently I walk through the kitchen, I open all the cabinets, even though I already know there's nothing there. So I'm really glad I don't live in an office that has like M&M's or something, because I'd be gaining weight like crazy. Thanks. And feel free to ask for me to elaborate on any of the other questions if you kind of have specific suggestions. For example, on the people ops 111 issue or things like that. Clement asks, how's the search for the CMO? Yeah, it's going well. We kicked it off. We hired a recruiter, Steve, from 343. And I really like how they're approaching it. It's been, it was kind of the best kickoff call we ever had. Based here. So I'm excited. But the bar's super high, so it's going to take probably months to find the right person. So yeah, personality traits. Thanks, Suri. That's a great question. Let me find some relevant materials. So this is like the experience and the characteristics we're looking for. I assume you can read faster than I can tell you about it. So I'll leave it for that. Yeah, Clement asks the story like what have we learned so far about CMOS and what are we looking for in the next person? I think one of the things that I'll be, two things I'll be watching closely, kind of planning back from a target. So in reference to the IPO, I already said it's much easier to kind of know where you want to go and plan towards that than knowing where you are and then guessing kind of what direction you have to go. And I see we need someone that does that, takes where we need to go and plans back from that and knows the implications for that, for like what they got to do today, this week, this month, and where they need the organization to go. And you need that because we're tripling every year and marketing kind of needs to go even faster than that. That's quite a pace. And if you're not diligent about how fast you have to grow, then you end up behind. Another thing I'd really love to see is a very strong kind of data and ROI driven decisions. In marketing, basically where we want to end up is that we look at LTV versus CAC per channel and then do investments based on that. Let me try to unpack that. So LTV is lifetime value. So with that, you look at not just what you the incremental ACV you get from a customer in the initial deal, but you look at like, how long are they going to be a customer? How much net retention is there going to be? How much extra are they going to spend with us? Are they because they could move to a higher tier or they add more users? And then the CAC is customer acquisition costs. So how much did it cost us to acquire? You could imagine you spend 100,000 bucks to acquire two customers. And both of them spend to have an initial order of $200,000. Great. Like your LTV-CAC ratio is two. Like you spend half to acquire them, what you get in the first year. What you want to get to is know how much they're going to be over time. If one of those customers you sold like to every user in that company and at the highest tier, and you got a one is a much bigger company where you just have an initial entrance, but there's so much room for expansion and that expansion actually probably will happen. That is a much better investment. Now you want to attribute that to the different channels and campaigns you're running. So know what is working. Is it content marketing that brought them in? Is it the Alba marketing that brought them in? Is it the analyst relations that brought them in? That's super hard to figure out, but it's very worthwhile because if you know what bought someone in, you can adequately kind of load those costs and you can see, oh, we're going to invest more here and less here or vice versa. Dan says, look, we tell customers they can release more often. Do we have a time frame for releasing more frequently on goodlot.com? It's improving, Dan. So it used to be that we did no release candidates. That's a long time ago. Then we did a few release candidates. I think the trend is more and more release candidates. We'll keep upping that to not just release candidates, but just making sure every single feature that gets merged into master gets released separately. Now, right now, we're still using our omnibus packages to release and we kind of have to build a new package for every single thing we do. We want to move off of that and towards cloud-native charts where we're using Kubernetes and it's much easier to do deployments, much easier to do canaries where you only send a certain part of your traffic to the new release. So we need better tooling and we're working on that. But we're not, as said before, with goodlot.com. We're not where we need to be, but the team is making progress towards that.