Want to go totally asynchronous? Repeat founder Sidharth Kakkar on building a remote team & autonomous culture

[00:00:00] Brett Berson So I I'd love to maybe start zooming out and, and having you share a little bit about the, the subscript, founding story.

[00:00:10] Sidharth Kakkar yeah.  subscript came out of, uh, my last company in a sense.  I made a lot of mistakes running that company, apologies to everyone who had to work for me. Um, and one of those mistakes was I, uh, didn't hire a finance team, um, really at any point.

[00:00:25] Uh, and then eventually, uh, when I started thinking about it, we ended up getting acquired. Um, and what that meant that I was defacto CFO, uh, apologies again to everyone who had to experience the other side of that. Um, but as a result, I ended up doing all the things that you want a great, uh, finance leader to do.

[00:00:43] Um, well I did 'em poorly, but I tried doing them, uh, specifically around using, um, financial data and information and collecting  your subscription information to make really good decisions. and I'm a pretty data oriented person. My team was a pretty data oriented team. Um, but we were sort of [00:01:00] very starved for high quality information and data to make the decisions off of.

[00:01:04] Um, and  it  felt like this big lost opportunity.  a source of unforced errors in that the data existed, we just couldn't get it collected and like processed fast enough. Um, and so we tried, you know, the usual path of, um, having lots of people work on it and having them spend a lot of time in spreadsheets, sort of doing really manual stuff, stuff that was honestly  Ray below their skillset.

[00:01:30] Um, and of course that leads to happiness for no one. And so that pain is what led, uh, me to think about. Well, do other people experience this problem and the answer turned out to be a resounding? Yes. And then is it the problem even solvable? Um, we think it is, and we've been working on it for a little bit and, um, Then, how do you go about actually making this much, much, much easier for finance teams, leadership teams and CEOs for others who are doing the thing [00:02:00] that I did, which is sort of playing defacto CFO themselves.

[00:02:03] So that's where subscript came from. And now we help, um, finance leaders at subscription companies get a much better handle on their data. Uh, get a much better handle on what's going on in the business and be able to actually advise their counterparts in sales or success or, um, even product and engineering on what's working, what isn't working, um, and down to the  sometimes even like  deal level discount decisions, all the way to prioritizing a particular segment versus another, um, to thinking about the ICP,  all of those things are things that we help with.

[00:02:38] Brett Berson given this was your second company, uh, How did you think about what are the pieces of your experience building freckle from either a way that you worked culture people, sort of all of the component parts, what are the pieces that you wanted to bring over when you started again versus wanted to rebuild?[00:03:00] 

[00:03:01] Sidharth Kakkar Uh, it's funny. So after, uh, freckle was acquired, I, um, I spent a bunch of time putting together this, like, 35 page doc that no one, but me has seen. Um, and it's just like this brain dump of everything that I can think of that I learned. And there's lessons that span everything from how to be a better leader, to how to be a better manager, to how to prioritize product better, to how to, um, hire better, um, to how to make tough decisions, uh, all the way to, how to build sort of a culture that does all of these things and grows and how to transition from the various stages of, um, what  the CEO job is.

[00:03:42]  and so  First, there was a sort of like the product and the space and the idea on subscript. And then there's sort of the company building side of things. And on the company building side of things, I spent a lot of time sort of re-reviewing all of those things.

[00:03:56] So  subscript is very much a reflection of sort of all the [00:04:00] things that I've learned, but also in a way that, you know, year six of running freckle is a reflection of the first five years.

[00:04:08] Brett Berson so can you talk about some of  those key learnings and how it applied specifically, maybe to the, the people and culture piece of, of subscript.

[00:04:19] Sidharth Kakkar Yeah.  So some, some things that are different from like ground up are subscript is a totally remote culture. Freckle made that transition. So freckle started out as a San Francisco only based team and in the office sort of five days a week.

[00:04:33] And it transitioned eventually to a hybrid. Um, there's a San Francisco team and then the rest of the us team, uh, we had only a couple of members outside the us, so largely it was in the us and, um,  I personally went from a person who believed, oh, you have to all be in an office and you want everyone to be together, 

[00:04:53] And now I'm very much on the,  other end of the spectrum. And, um, I'm, I've sort of fully [00:05:00] embrace the full remote, but then we started thinking a little bit more about, okay, so if you're gonna go fully remote, the next thing you could do is not have to have everyone kind of work at the same times, because that allows everyone to kind of do things in their own way.

[00:05:15] And I personally find this. Uh, quite a bit where there's just times of the day where I am much less productive than other times in the day. Uh, one of our earliest engineers, Brandon likes to dance after lunch. And so he, he just spends a bunch of time dancing and then he comes back to work refreshed, 

[00:05:33] In a much better place and that's sort of what you want, right? You don't want people sitting in front of their computer when they're not gonna be able to be productive. So that was  the second part we layered on. Well, you can work anytime you want, because you want you to work when you're most productive, but then you have this question of  how do you do meetings?

[00:05:48] And so the next step will, let's just not do them. So let's get rid of meetings. Um, and so subscript is a zero meetings, culture, uh, which means that we have zero internal meetings. Um, the only exception to that [00:06:00] is meetings for fun. So we'll do like board games together, or  team fun activities together, but that is it, uh, zero other kinds of meetings.

[00:06:08] Um, and then if there's doing zero meetings, then you sort of go, okay, well, do you have to keep everyone in the us? Um, and you have this realization you don't because, um, the person working in Asia no longer needs to wake up in the middle of the night to come to an all hands anymore. and so we went fully globally remote.

[00:06:28] Um, and so now we have this culture that has fully globally remote and has no meetings, uh, and everyone can work whenever they want. Uh, and  I can get into a bit more about my, my opinions on micromanagement later. Uh, but there's this very sort of full autonomy culture then layered on top of it, because if you have people working whenever they want, that means that other people who, if they need someone else to make a decision for them is probably not working when they're working.

[00:06:55] Right. So, uh, we've emphasized that everyone gets to make their own decisions as a [00:07:00] result of that. And so from that follows, well, if everyone can make their own decisions, then you have to make sure that you're hiring people who are going to be able to make those decisions. And so that changes our hiring philosophy, a decent bit.

[00:07:11] And then if you're gonna hire people who can make their own decisions and empower them to make their own decisions, well, then you have to make sure that they have all the context they need cuz in a traditional organization, often  the reason people aren't allowed to make the decisions is because they don't know all the things they need to know.

[00:07:27] And so then we go really hard on the context side of it, but that means now you have to have a large amount of transparency. And so as a result, we have an extremely transparent culture where everyone can sort of know everything. Um, and so  all these pillars kind of follow each other, you know?

[00:07:44] Um, and what's really nice is when you're starting out a new company and you've kind of done this before, you can look at that story and you can see how all those pillars follow each other. And so from day zero, all these things were true about subscript.

[00:07:58] Brett Berson What was your process [00:08:00] in landing on this? What was sort of the reflection process, I guess of looking back at freckle and, and sort of figuring out the bits that worked and maybe the bits that could be improved.

[00:08:13] Sidharth Kakkar There was a lot of, sort of my own evolution as a CEO, um, EF freckle, uh, being a first time founder CEO  it's a very tough gig. I, um, really empathize with everyone doing it because there's so much you need to learn. And at the same time you have to like be certain enough in it and confident enough in it to like rally a bunch of people around it, whether it's, you know, your customers or your investors or your employees, but at the same time experience has value and you don't have it.

[00:08:42] Um, and so you kind of have to like, Gro around in the dark a little bit and try to  find the right things to go after, uh, or the right beliefs to have almost. Um, and then there's the fact that there's  multiple right answers, right? Like I just landed on an answer that's right for me [00:09:00] for now.

[00:09:00] Right. I will continue to evolve. Uh, but there's people who believe the exact opposite of me and  many of them are much smarter than me. Um, and I don't think one of us is right, and one of us is wrong. And so there's multiple answers. That means there's multiple sides of advice  and you kind of have to sort through it.

[00:09:15]  And for me, that meant  there was a whole lot of evolution that happened in the six years, um, that I was building freckle sort of as an independent company and even in the year and a half where I was, uh, building it  within my acquirer.

[00:09:28] And one of those ways that I evolved quite a bit was how I manage people. So I started out in a, the way a lot of people start out as managers, which is as fairly micromanaging. I didn't call myself that at that point, if you had asked me if I'm fairly micromanaging, I would've scuffed and said absolutely not.

[00:09:46] Um, but  in reflection, I did a lot of things that I would now consider quite micromanaging. Um, I do now use this term way too liberally though. And I will sort of like call almost everything micromanagement. So like to take an extreme [00:10:00] example, um, our engineering team doesn't do a traditional standup, you know, one where it says like, what did you do yesterday?

[00:10:05] What are you gonna do today? Uh, and do you have any blockers? I only find one of those questions helpful. I find the first two to be quite micromanaging. Um, and so we only talk about blockers, but there's no, like, what did you do yesterday? And like, why didn't you do more? Like, are you doing it? Like, we don't do any of that type of stuff.

[00:10:22]  there's very minimal benefits to micromanaging people. Um, but there's quite a lot of downside to it. Um, so that's sort of one example of like a thing I learned that like played heavily into the autonomy based culture that we're building 

[00:10:35] Brett Berson you reflected on what was working or what didn't did, did you spend time contemplating the idea of did this thing work because of the thing that I did or in spite of the thing that I did.  when you think about very complex organisms like startups, I, I think that is one thing that is underappreciated.

[00:10:59] Right. In the [00:11:00] sense that, you know, a founder chooses to have  no managers for the first four years of the company. And so the head of engineering 65 direct reports and the company was successful. And so it's easy to say, okay, the next time I build a company I'm having no, uh, managers until five years in, 

[00:11:20]  and maybe  the company was actually successful in spite of that philosophy  do you meditate on that in any ways you thought about what you want to do in this new company in terms of how you operate and behave.

[00:11:33] Sidharth Kakkar  I feel like you constantly relearn this idea that life is almost always very nuanced and the like one way or the other answer is usually wrong. And it's funny, cuz I still though in this case have a,  not perhaps as nuanced as it should be type of answer. Um, I think that the primary determinant in success of a startup more than pretty much everything else is, [00:12:00] um, the strength of uh, the initial product market fit.

[00:12:03] I think everything else improves or, or like takes away from that. But I don't know, like product market fit really does seem to solve everything. And  I'm sure we all have come across companies that seemed and sounded extremely poorly run 

[00:12:16]  and the company still end up being super successful. And I just feel like product market fit  solves a lot and can hide a lot.

[00:12:24] Um, so as I approach this, I very much of the belief that the, the level and quality of the product market fit, you have ends up being primary. And then everything else is a little bit of like, well, what kind of culture do you want to work in? And it's, it's one of the nicest things about being a founder. You  get to work in the kind of company you want to work in because you know, you made it

[00:12:48] my biggest sort of recommendation would be pick a coherent set of things.

[00:12:54]  there's a lot of your answers to this, but it definitely helps to have coherence here. If you have, um, a [00:13:00] set of philosophies that are a bit of this and a bit of that, like if you try to do what we do with the no meetings thing, but then you're in different time zones or whatever, you're gonna start defaulting back to meetings.

[00:13:10] You're gonna be like, oh, why isn't this working

[00:13:11] Brett Berson Do do you think that the choice that you made in terms of a sync remote, et cetera, will increase the likelihood that you get to really strong product market fit or to the point that you just made? It's just more your own working style and preferences.

[00:13:31] Sidharth Kakkar  So I think, yes, but I will say that I am not super confident in this opinion. I, I could easily be convinced that, um, there's advantages to helping get to product market fit and other styles. The reason that I think yes for this style is for example, in our product market fit journey.

[00:13:47] Um, there's no, Pretending anything in our company, right? this is a thing that  founders or leaders often do where the real truth, the real unvarnished truth, like sits amongst them.  And then they [00:14:00] think about like, oh, how do I massage this message?

[00:14:01] The rest of the company? There's none of that at subscript. Right? Because everything is public. So sort of everyone gets the full unvarnished truth. And what's, what's amazing about that is like, everyone's always afraid of like, well, what will people think  if they realize that, you know, on the product market fit survey, we're currently only at like a 30% 

[00:14:20]  like, will they lose heart or whatever? I, I just think that if you treat people like adults they'll act like adults, and if you don't treat them like adults, well, you're gonna get what you asked for. Um, and so we have this culture of like, everyone knows exactly where in our product market fit journey.

[00:14:34] Everyone knows  all the things that are. Positive developments in that journey and negative development developments in that journey.  But as a result, one, we, we sort of communicate to the whole company. Like  you're a huge part of this journey, just like the everyone else here and that you have significant things to contribute to it.

[00:14:54] And we trust you with this  information that's sometimes difficult  And. What we really [00:15:00] want here is we all want us all to work towards this  very clear goal of product market fit. And I'm not going to transform that or like turn it into something more localized for you, right?

[00:15:12] Like I'm not gonna go over working towards product market fit. And what I really want, um, Emily on my team to do is X, Y, Z. I'm not gonna do that. I'm basically gonna go to Emily and say, Hey, we're working towards product market fit. And here's how we'll know we get there. Uh, and here's sort of all the data that's available.

[00:15:30] Please figure out what you think makes sense in your role to help the whole team get towards product market fit. Um, and so I, I think Subscript is more of a, of like a collective brain rather than, um, a traditional organizations where like maybe just  me and Michelle would know all the things and then everyone else would get like, you know, processed information here.

[00:15:52]  and everyone's, um, exposed to all the decisions  being made across the organization and they can contribute wherever they feel like they have useful things to [00:16:00] contribute.

[00:16:00] Brett Berson before we get into to the sort of how you're executing on it, I, I, I'm interested to talk a little bit about what maybe someone's gut instinct  or immediate reaction would be when they th when they hear remote and ay, and I think, you know, you could probably cluster it into a few, but one would be.

[00:16:19] so much of, of working at a company and specifically a startup is about these deep relationships that you have with coworkers and friendships that you develop with coworkers. And what, what happens to that in this context? And do you, do you ultimately sort of just not think that it matters or you're able to nurture that thing in, in the way that you're building the company?

[00:16:52] Sidharth Kakkar If you were to take our slack and sort of the conversations that happen in that slack, um, many of them, [00:17:00] you know, asynchronously in that, like responses are come a few hours later kind of thing.  I don't think you would feel like, oh, this company culture is different than the culture of a company.

[00:17:10] That's all in person. the emojis would like feel natural. and then you would feel like the banter, um, that exists in channels, like our food channel or like our parents channel, uh, is quite a lot of parents at, at subscript, not shockingly this type of approach, appeals to parents quite a bit, um, where they can have actual control over their schedule.

[00:17:31]  And then when we do our sort of like, uh, monthly, uh, we call it live time, um, for an hour, an hour and a half.

[00:17:38] Um, and we,  do like, get to know each other activities or like, um, board games or stuff like that, it feels just like any other company. It does not feel stiff or awkward in any way. 

[00:17:50] Sidharth Kakkar  On top of that, a couple of other things that we do are, you know, some of, sometimes these like live sessions involve, um, uh, homework and the homework is like, you spend [00:18:00] an hour with, uh, you know, randomly assigned  teammate, uh, in the company.  we did this one interview of like favorite toys from your childhood.

[00:18:07] And we had like all these pictures of, to toys and stuff, uh, or places that you lived when you were growing up. And so  we get to know each other in all these, um, deeper levels. Uh, and then finally, we also do, um, we do in person offsite. We, we're still trying to figure out what the right cadences given.

[00:18:25] Uh, COVID  Our goal is three or four a year. And, um, and we spend a whole week together, uh, where we just bring everyone together to the same place. And, um, we do, you know, mix of the usual offsite stuff like working together, brainstorming together, culture norms together, having fun together, all that stuff.

[00:18:42] So we have like a bunch of different ways that we  fill that hole of like being close to your teammates. And we, we, like, I feel like most people feel quite fulfilled around that. Um, and I do too. And, um, my, you know, Michelle is like an extremely [00:19:00] social person and she feels quite fulfilled around that.

[00:19:02] I feel quite good about this. Um, I feel like it works really well.

[00:19:07] Brett Berson So, what are your observations in terms of  the key downsides to this methodology thus far, or sort of the explicit trade offs that you've been willing to make?

[00:19:21] Sidharth Kakkar Yeah, I think maybe the, the biggest thing is,  at an extreme, if you are the kind of person who just has to be in an office every day, surrounded by your coworkers in order to feel excited, motivated, whatever, then this doesn't work for you.  the social part is totally there, but it's not there every hour of every day.

[00:19:43] Right. Um, you're not gonna have lunch with your coworkers  So if, if like that is really important to you, then this doesn't work. But I generally have this belief that hiring is always, uh, segmentation, exercise, just like finding a market for your product [00:20:00] is a segmentation exercise, but what's even nicer about hiring is that you can really micro segment.

[00:20:05] Like you basically can find a very small segment because  you're not looking for a big Tam when you're hiring. You're looking for a very tiny,  few people, uh, but you just need there to be overlap. I mean, the people you need and the people who are interested in, um, in the sort of product you're offering or the product you're offering is the culture that we they'll work in.

[00:20:24] And. I feel like basically this gives us  a very specific segmentation. Um, we hire people who really care about, um, managing their own schedule, who are very good at motivating themselves, um, who are very comfortable making their own decisions. And ambiguity, uh, were very good at,  communicating in a written asynchronous way.

[00:20:51] Communicating a synchronously via video and who, uh, have, uh, sort of the, the type of skill sets that we're looking for. Um, and the [00:21:00] things we don't need is we don't need anyone who lives in San Francisco is willing to move to San Francisco.  we don't need someone who, uh, like, uh, has to be in an office from,  nine to five or whatever.

[00:21:13] We don't need someone who has to wake up for a all hand at a particular, like, we don't need any of those things. So basically we're like we picked our segment and, um, the it's both a  positive and a negative, I guess. Um, the negative is that, uh, we can't hire people outta that segment, but I think that's mostly just a positive in that we have a very specific type of person, um, who really loves

[00:21:38] Brett Berson with that as a little bit of a backdrop, let's talk about how you would describe  the component parts or the operating stack or rituals, um, to make this happen.

[00:21:55] Sidharth Kakkar  a lot of this also came outta learnings from my last company. So [00:22:00] we eventually developed a pretty good documentation culture at freckle, but we didn't have one on day zero. Um, at subscript, we had one on day zero. Uh, everything gets documented at subscript. Um, and we try to make sure that. All the logic of every decision we make is reflected  in our sort of collective Wiki  like EV everything, even the smallest things.

[00:22:24] Right. And there's a preserved record of the discussion that was surround in it because there wasn't a meeting where people discussed, uh, what we should do. There's a written record with comments  and sort of like evolution of thought that all gets written. And so the written culture is a huge part and I think should be a part, whether you're synchronous or a synchronous.

[00:22:44] In my view, uh, this, this becomes, you know, at, in year six of freckle, there were so many things that we basically ended up re-litigate that we had figured out like three years ago, because. It's a lot of different teammates and, uh, you forget what you decided, you forget why you decided [00:23:00] you're asking the same questions again.

[00:23:01] It's so good to have a good documentation culture, um, early. Uh, so we did that on day zero, but also it's for us, it's not even optional in that. Um, all, all of discussions are in writing. Um, and the one thing we did also differently is, um, in most like most companies at fr Africa, we used a lot of Google docs.

[00:23:19] Uh, we don't use that here. Um, it's it's for like a reason that sounds trivial, but it's really not. Um, and the reason is that. In Google, you can create a doc, uh, and you can share it with someone that doesn't live in any folder or hierarchy. Uh, and that means a lot of things basically go into the ether, uh, and you will never find it again.

[00:23:39]  there's lots of, uh, products you can use. Uh, we use notion last time you used confluence, both are great. Uh, but basically the main thing that they do is they enforce a hierarchy. 

[00:23:48] and so you can always find things easily. And that's such an important thing about sort of the, the documentation heavy, uh, culture, the other big, big lesson that plays a really big, uh, factor in [00:24:00] this is like how to do really good goal. So most companies, I think go through many iterations of goal setting.

[00:24:06] And, uh, it's interesting. OKRs are a tool that are very simple sounding, uh, but actually executing on them can be quite hard and nuanced. Um, cuz there's all sorts of  little things, uh, about how you frame your OS, how you frame your OKRs and um, like how outcome based you are versus how, um, sort of process based you, all these things matter a lot.

[00:24:26] Um, and I, I also think that this is another one of those things where there's a lot of right answers, but you have to find what's right for you. And mostly it just takes experience to do that. Uh, so luckily I have  the entire experience of freckle to draw upon here. Um, and so at this point I have like a pretty strong philosophy on OKRs, 

[00:24:42] And so that philosophy is sort of infused across how we, um, sort of. Set the context for what we're doing. Um, so a part of our kind of operating cadence is we have monthly OKRs. Um, so right now we're working through our June OKRs. Um, [00:25:00] and they're very simple, usually only one or sometimes two objectives.

[00:25:04] And then like two to three KRS per objective, uh, because focus is really, really important. And, uh, so we're focused on two things in June. Um, and off those two things,  every week  usually me, but sometimes other members of the team will do a context setting video for that week.

[00:25:22] uh, so this is something that goes out, uh, every Sunday night, uh, PST. So like Monday morning for a bunch of people, uh, who start in Asia. Um, and this, this video essentially, like here's what our goals are. Here's our KRS, here's where we are on those KRS roughly right now. Um, and then we go down to,  here's like, uh, a set of ideas for this week that can probably help us move towards those KRS.

[00:25:48] So everyone gets this level of context. Right. Um, and that the set of ideas isn't like. do X, Y, Z. It's more like, um, problems a or B or C are  the [00:26:00] types of things that are standing in our way. Uh, and then occasionally, if there's like a really big project, that'll be like, okay. On like making progress on these projects will probably help.

[00:26:08] Um, so that's like, you know, five, six lines of, of  context and, and bulleted format. Um, and then the last part of this weekly, um, video is reflections from the previous week. So everyone collects all the reflections at the end of each week, uh, in the doc from that week. Uh, and then  in the video, um, we pick a few of the, uh, reflections that it, it makes most sense for sort of everyone to, to have context on.

[00:26:34] Um, so these reflections can be anything from like an awesome win that we had, um, that like is just a great thing to celebrate together. Uh, or it can be like a, oh, like someone saying I'm really. Um, I'm really drowning in notifications. Right. Which is a problem that almost everyone has. Uh, and then sort of like, uh, using that as a prompt for a more, um, a more in depth discussion that all obviously also happens [00:27:00] asynchronously on like, how should we think about, um, managing notification sort of, uh, overload.

[00:27:05] Um, and that video is like maybe 15 ish minutes and that's our replacement for all hands. Um, it's weekly, uh, and, uh, it's short, um, people comment on it, uh, and people comment on the doc. Um, and so there's robust discussion on it. And so like, it's just in my view, a significantly more powerful thing, uh, than an all hands, um, because you can do it way more frequently, which I think is quite important.

[00:27:29] All hands are very important in my view. Um, and you can keep it much shorter, um, and you can get really good back and forth on it.

[00:27:41] Brett Berson there's a lot to sort of dig in from what you just shared. Maybe we could go back to the, the documentation culture. I'd be curious to learn more about  what is the hierarchy in terms of how you organize this information? And I think number one challenge is always, you know, you have hundreds and hundreds of documents [00:28:00] or thousands over time.

[00:28:02] Um, how does someone find something? How do things get sorted? Um, and so I'm interested. What have you figured out, or how have you designed the, the, this sort of, I, I guess in this case, very pragmatically, your notion, you know, if I were to go in, what would I see? How is it organized and maybe how did you land on that?

[00:28:22] Sidharth Kakkar 

[00:28:22] Yeah. So there's a few main sections. Um, the, the first one is overall company level and here generally there's, there's not much, information's sort of like, you know, what's our company address and stuff like that. Um, but the main sort of functional sections are there's the goals and planning, which, uh, I just talked about.

[00:28:39] That's what we use to actually do, like kind of our, all hands for the, for the week or our OKRs for the month. Uh, and  almost all these sections have like a discussions, um, area where it's sort of a, you can dump any doc you want in there that pertains to that area. So if you were having a discussion on like, is there OKR process working and in what ways should be it [00:29:00] Arian on it, it'll go in that discussion section.

[00:29:02]  then there's product and design.  that one also has, uh, RFCs section, uh, just requests for comments. Um, and there's a lot of those. And we specifically name all of these things, request or comments.

[00:29:12]  I really like it because it sort of like very explicitly says, this is not decisions or something of that sort. This is, um, a place where I'm looking for input from other people. um, and that in product, we also have things like discovery interviews or feature requests or, um, and then there's, uh, within there's one of my favorite things, which is what we call our journal of product market fit.

[00:29:35] Um, so in that journal, there's essentially like entries that are, uh, at a very high level, what is working towards our product market fit and what are the things that we need to think through? Um, and it's very fun to read because basically we started doing this in like month two of the company.

[00:29:51]  And, uh, you know, some of the other ear early, like discussions in there, uh, are called things like, um, you know, lighthouse, customer [00:30:00] problems and solutions, or like. Uh, for whom are rebuilding, uh, things like that. Right? So like very high level, like Hmm. Where we, what should we think about?

[00:30:08] Uh, and then more recent discussions are more like, uh, our core buyer, which is the finance leaders and other stakeholders and how they work together. And what does that mean for our journey in product market fit? Um, so this is, this journal has been like one of the coolest things, cuz you can watch our, um, thinking and our business evolve.

[00:30:26] And there's usually like, I'd say like four or five entries a month in here. So there's quite a bit now. Uh, in fact, June has six entries already.  so like we are getting a lot of really good, hard thinking done. Um, and then the next sort of high level section is engineering.

[00:30:42] A lot of discussions happen here. So tons and tons of RFCs, every sort of big, uh, or small engineering decision that gets made. That's not at the level of a code review happens. In notion. Um, so if something's like,  a comment someone left in my code review recently is like, I [00:31:00] used a lot of string literals where I should be using constants.

[00:31:03] Uh, and that was like a mistake I made, cuz I didn't realize that was all set up. Um, so that happens, not here, but for anything that sort of like,  how should our NetSuite integration work? Um, because NetSuite is quite hard to integrate with, um, that would show up in the discussions on the engineering section.

[00:31:19] Um, then the go to market section, mostly just discussions, but also a lot of like any document that we share with a customer, um, tools that we're thinking about using, um, if we have goals for that functional area will go there. Um, and like how we thought about those goals. So,  if we're ever like, oh, why was.

[00:31:39] May goal X, uh, you can actually just go, you know, exactly where in the hierarchy could to go to see exactly what the may goal was and why it was that and how we did against it. Um, and this is  history that will be here forever. Right? Um, we'll, you know, six years, hence or 10 years, hence, uh, we'll be able to see, oh, in may 20, 22, if you were curious, uh, what things [00:32:00] look like.

[00:32:01] Then there's  a client success section, uh, and here there's everything from a lot of how tos, um, to like, uh, more RFCs, um, notes from specific customer calls, stuff like that, that all goes here as well. Um, so we actually do keep, um, sort of our, uh, running log of how specific customer relationships are going in here.

[00:32:24] Um, we have a section on culture and another one on team. Um, so on the team, one, it's more about like hiring and we have a sort of subscript hiring guide, um, and this continues to get improved. So every time someone makes a higher subscript, we say like, oh, can you go back and add things? You've learned in the process back to the hiring guide.

[00:32:43] So we like do a lot. Continuous iteration on all of these things, um, explicit and there is our compensation philosophy. So no one is ever wondering how, uh, their compensation came about. There's a very explicit philosophy that you can read. Um, and it's right in there. Uh, [00:33:00] our buddy program documentation is in there, um, on the culture side there's things like, uh, our core principles and, um, what we call our guide to being a stellar subby, um, and things like that.

[00:33:13] Uh, and then there's like a general operations and a security section and, um, stuff like that.

[00:33:22] Brett Berson And so when you say, know, many of these, uh, top level sections have lots of areas, that discussions are happening, it, it, what does that actually look like? Is that just people commenting back and forth on a document that somebody puts together, or what is the discussion format?

[00:33:39] Sidharth Kakkar Yeah. The way that'll typically happen is say,  Emily who runs client success is thinking about, uh, changing the process a little bit for how, um, how client success starts to become involved in a customer relationship during the, the sales and the data set phase. Right. Um, and so what [00:34:00] she might do is like, uh, have a little bit of background on like, why she's thinking about this and, uh, maybe a little bit of additional reading, maybe like things she found on the first round review of course, or things like that.

[00:34:12] Um, and that sort of goes into like, here's some interesting reading on it. Um, and then how things are currently working, what's not working about it. And then like what she thinks might work a little better. Right. And very much every discussion is framed as like here's some ideas I have, I would love to hear opinions.

[00:34:28] Right. Nothing is like, sort of, here's a decision I've made. It's, uh, here's some ideas I have. To be clear, Emily's the decision maker on this. And, uh, no one else gets to make the decision. And so we're like really big on, uh, each person being sort of the directly responsible individual for a particular idea or a task.

[00:34:45] Um, so it's still one person, uh, but it starts out with, I'm looking for input from others on this. Um, and then in that doc, she might like specifically tag the people who, um, need to work with her on this. 

[00:34:58]  and those people [00:35:00] then sort of the expectation is within the next day or so. Um, they'll have a chance to go back in. And actually leave comments on the things they agree with, the com things they don't agree with. And then, um, often what people will do is add a section at the bottom, write their name and say like, here's some  other thoughts I have.

[00:35:15] And here's, uh, a sort of slightly different idea I have on how we could do this. And then, uh, from there, the discussion will sort of naturally evolve as it might in, you know, any discussion except it'll be like multi threaded and all, any synchronous. Um, so you get in my view a much richer discussion and a much more clear sort of, um, communication because.

[00:35:37] If you misread something, you can ask more questions and then if other people also misread it, then they can also see the explanation. Um, and you can go, you sort of have a chance to like reread things.  and then from there, like at some point,  all discussions tend to converge into some like clear solution or next step.

[00:35:55] And that's where,  Emily, in this case would be like,  as a directly responsible [00:36:00] individual, the DRI on this, um, I think we're gonna do X, Y, Z. Um, and you know, implicit in there is like, she's now got discussion and,  at least. Enough buy in from the other people who need to be involved in this, um, that that becomes sort of the decision.

[00:36:15] And now that's sort of preserved forever. Um, and what's really cool is that let's say a new client success person joins and they're like, wait, why do we do it this way? You can just be like, oh, check out this doc. And if you have new ideas on it, just create a new section at the top, uh, or create a new doc, reference this one, uh, and then we could continue iterating on it.

[00:36:33] Um, and so  you're never like re-litigating decisions cuz you can just see everything that was involved and all the considerations that went into it, uh, rather than just like making people repeat themselves. 

[00:36:44] Brett Berson So why do you think I, I, this sort of methodology tends to be atypical, where if you look at the average sort of scaling it tends to be meeting centric. It tends to be more verbal or real time [00:37:00] versus async versus a sync.

[00:37:03] Sidharth Kakkar I think it's cuz it's easier. So if I'm in an office with a bunch of people, this is sort of like the much wanted, um, run in and like, uh, serendipitous conversation that different people can have kind of thing,  if I'm in an office and I run into people and then we have this conversation, or maybe we have it over lunch or whatever.

[00:37:23] Yes. You get the serendipity, but you've also now created a bunch of knowledge between the two of you that, uh, no one else is privy to. And maybe some version of it gets documented, but likely what's gonna happen is, um, maybe you'll change your behavior in some way and start doing things a slightly different way.

[00:37:39] Or maybe you'll  process the information, decide something, and then let everyone know what you decided, but now they don't have enough the context. And so I guess like, um, maybe, you know, an explicit downside of subscripts culture is like we're missing the, um, random serendipitous interaction.

[00:37:56] Uh, but in, uh, I guess from a very empirical [00:38:00] level, I honestly have not noticed the downside of it. It seems fine. What we've gained though, is this  incredible written permanent, um, sort of well documented technology that has made so many things so much easier

[00:38:17] Brett Berson So I wanted to loop back to your goal setting process. And I'm, and I'm interested kind of given this orientation around extreme autonomy. Can you walk us through the process by which you mentioned you have one or two objectives with one or two that you're focused on, on a monthly basis. And then how does that map to what a software engineer a data engineer or a product designer is actually doing on an hour by hour, day, by day basis throughout the month?

[00:38:53] Sidharth Kakkar so  this is a real example, uh, for us  would be, um, we wanna make sure that, uh, every [00:39:00] customer, every time they log into subscript, uh, can 100% trust everything that they see, right? Like there's never a, um, sort of, um, questioning the data and that sort of has a lot of things encapsulated in it, but everything should work every single time.

[00:39:16] And that's, that's the thing that we that's the bar that we hold ourselves to. Right. And so that's, that was one of the objectives actually, uh, in last month that we wanna work closer to getting to that bar. Um, and within that, the KR, so as a bar, the way that came to be was actually based on a bunch of discussions in the journal of product market fit, where we sort of evaluated what's working and what's not working in April and before.

[00:39:42] And one of the conclusions that a lot of those documents pointed to was, um, that, you know, this is a place where we need to improve. Um, and the first thing I'll usually do, um, is a few days before the end of the month I'll, um, post here's the objective, um, that we're thinking about, [00:40:00] uh, for the next month.

[00:40:01] Um, and so that's like sort of open for comment, um, and because we've already had all these other discussions that have led to that objective, usually in the journal of product market fit, um, everyone kind of is already aligned around it.  most of the time everyone agrees that that's sort of the obvious objective.

[00:40:17] Um, so that's awesome. And then for the specific KRS,  Usually that's a little bit more of a collaborative effort. Um, and so I'll ask a couple of people who might really have good ideas on how to, um, make sure that we're measuring this properly. Um, and then there'll be  other folks who'll have other ideas, so there'll be a little bit of back and forth.

[00:40:35] But, um, we are not, we're not like trying to turn, you know, the K generation into a, um, super intensive process.  So, uh, we essentially, you know, pick cares that seem pretty good. And, and, you know, we just roll with it.

[00:40:49] Uh, the point really in my view of OKRs is to get everyone very aligned, um, on what we're all working on. And, uh, everyone directionally [00:41:00] moving towards what is a good level of the thing that we're trying to accomplish. Right. Um, and so we can get that without, you know, trying to be perfect on the KRS. 

[00:41:08] So we have that. And then on the weekly videos, Essentially, we talk a little bit about, like, for example, um, if I look at one of the, the may weekly videos, um, there was, uh, like three or four things in that one. Um, one is like, um, This is towards the end of may. We had a lot of sort of things already in progress.

[00:41:30] So there was little less to talk about. So it was like, oh, you know, making progress on all the things that, um, have a data integrity label on GitHub is what a, uh, engineer might end up focusing on. Uh, and then, so then the question is like, well, how does that label get there? Um, and it essentially gets there by, um, because everyone is aligned on, this is the goal that we have.

[00:41:50] Um, and then based on that goal, people have  ideas of projects. So that shows up in, um, either an engineering or. C or a product RFC of [00:42:00] like, oh, here's ways we could improve data integrity. And then some of those, um, you know, really, if they really speak to an engineer, for example, uh, so actually a really good example, uh, on exactly this point is, um, there's a very cool idea that one of our, uh, principal engineers on had on how we can make our data integrity significantly better by adding this  new kind of, um, data checking framework, essentially.

[00:42:26] Um, and he sort of wrote up an RFC, um, like three or four other engineers, uh, commented on it, refined it, refined it, uh, got input from the data team, which added more ideas to it. And then, um, on sort of becomes the DRI for this project. Uh, so he, he came up with a he's the DRI for it, it's aligned to our goals.

[00:42:47] Um, and then he created all the stuff in GitHub. He added the labels to it. Um, and then, um, At this point, I think three or four different engineers have picked up and decided to work on, uh, [00:43:00] tickets related to that specific project. Um, so that that's like sort of a little bit of how the pipeline here might work.

[00:43:06] Uh, super autonomy focused. Everyone gets to make their own decisions. Everyone's choosing what to work on. Uh, but everyone has more than enough context to help them make the right decision.

[00:43:18] Brett Berson you changed this process since starting the company? Or did you of set out with this overall the way in which goals and product planning and execution kind of fit together? 

[00:43:31] Sidharth Kakkar 

[00:43:31]  So the process has evolved a little bit, um, but  the major contours of it have remained roughly the same.

[00:43:37] Um, the ways that it's evolved a little bit is that. I in the beginning, I was not quite as good at achieving the thing I wanted, which is I want this to be a fully sort of autonomous culture where the, the process I described to you, where on came up like a really cool idea that helps us accomplish our goal, which I there's no way I would've ever figured that out, but this is sort of like the collective subscript brain, [00:44:00] uh, going so much further than any one person can.

[00:44:03] Um, and, uh, that is what we wanted. But in the beginning it was a little bit harder. I think, partly it was like, when folks are relatively newer at subscript,  they have a little less context of the history, uh, of all the things. And also  it can be a little bit, uh, of a transition to like start operating in this way.

[00:44:21] It's like incredibly freeing. Once they've made that transition, but at first it can feel a little uncomfortable. Like, what do you mean I'm not assigned a ticket? Like, what do you mean I can just do anything, uh, that, that can be  hard, um, when you've spent, you know, uh, a decade or two doing the opposite.

[00:44:39] Um, so there's a bit of that. Uh, but I think what I've really focused on for myself is each time someone doesn't know what to work on or each time, um, someone works on something that lies like clearly not aligned. I ask myself what is sort of the bug in the system that I should go fix. I'm an engineer.

[00:44:58] This is how I think,  everything that needs [00:45:00] fixing is a bug. And so like what's the bug in the system and the bug in the system is always something related to like what I did to set context. So, this sort of comes back to  my beliefs around micromanagement.

[00:45:11] I think anytime you're telling someone what to do, um, you're not actually solving the bug. Um, you're like, I don't know, doing  a really bad patch and, you know, taking on more tech debt or culture debt or whatever. Um, instead you have to think about like, what are the system level things I should do differently so that, that bad decision or, or like misaligned decision doesn't happen.

[00:45:36] Brett Berson  So I, I was hoping you could expand on  this idea that micromanagement is bad or a bug in the system and maybe a little bit more detail, and we could, um, get more of your

[00:45:52] Sidharth Kakkar Yeah. Um, I tend to speak in a bit of an extreme way on this one. Um, and the reason I do it is not because I think that the, the [00:46:00] topic is actually black and white or like, um, always one white, but mostly because I think, um, almost everyone makes. The same mistake in one direction. And so I, I, um, find it helpful to speak very forcefully in the other direction.

[00:46:14] So the mistake that everyone makes and I made for much of my career, managing people is doing a bit too much telling them what to do. Um, and sometimes that's like very specific telling them what to do. Like sometimes it's sort of like, um, I dunno, anyone who's ever worked at a consulting firm investment bank will tell you, there's a lot of like telling me what to do and that like highlight this cell, this color and like italicize this type of thing.

[00:46:36] Like, I mean, there's like that level of telling them what to do, but then there's like sort of more, uh, in things that seem not as offensive, but are still, uh, in my view, quite micromanaging. Um, so an example could be, let's say you are, uh, this a startup CEO and, um, a salesperson who is working on a deal.

[00:46:57] And they're taking a particular [00:47:00] approach to like how to answer a particular set of questions from, um, the, the like customer over email, right? Like a lot of, um, startup CEOs who have done this selling before might be like very tempted to say, no, no, no, here's how you should answer this. Or like, oh, you shouldn't use those words and use these words instead.

[00:47:17] Or, um, like here's something that I would do differently  my sort of recommendation to everyone is to do less. Um, like even if you think someone's making a mistake, just let them, um, for two reasons, uh, the first one is, um, a lot of the time you think that they're making a mistake, they're actually not.

[00:47:36] And, uh, what you have is not the correct answer, what you have is just an opinion. Um, and so by. Um, forcing your opinion upon them. Um, you're essentially one potentially getting a better answer.  you're just happen to be wrong. Um, I dunno, I, as CEO am wrong very, very often and I, I don't think that's unique to me as a startup founder.

[00:47:59]  [00:48:00] and then the second reason is even if you are right, um, there's a question of how right you are, uh, because there's a cost to telling someone what to do. Uh, you've sort of like squashed their creativity a little bit. You've squashed their ownership a little bit off the thing that they're working on.

[00:48:16] Um, and you were sort of setting this pattern of like, uh, if I don't do things, uh, sad art's way, then that means that, um, he's gonna. Think less of me  and so I better just like run all my decisions by Sadar. Uh, and that is the worst possible outcome that you could imagine.

[00:48:35] Right? Like now you're basically like, um, in a collective company brain, you've, you've shut off a node. 

[00:48:41] so. You better be like that much more right. To like incur that level of cost if you're gonna do it. And then the last part is even if the decision is like just super freaking wrong, uh, but is reversible or at least is like a thing that, um, has a cost that [00:49:00] you can bear, um, your, your best off, just bearing it and then using it as a way to like, help teach that person.

[00:49:07] Uh, because in an ideal world, you don't want to be the person who's like sort of orchestrating what each person is doing you want a world where each part of your company, everyone can make decisions that. Correct that are great. Right. Uh, and that are culturally consistent that are consistent with their goals and so on.

[00:49:27] And so you want to get everyone to a point where they're able to do that and the way to do that, isn't to like, um, shut off an experiment that they're doing in, in like sort of their way of doing it, uh, and sort of replacing it with how you would do it. It's to let them run that experiment, let them try it their way.

[00:49:45] And then to like, Close the loop on, did it work? Why did it work? Why did it not work? And then help them if, if you had the right answer, uh, and you just had the, the totally right context, then going back and saying like, okay, well, here's [00:50:00] some things that I would've thought about differently in that situation.

[00:50:03] And I don't know if that would've resulted in the right answer, but it maybe is more likely to, uh, or maybe would've been more successful. Um, and I've definitely like, I am certainly not perfect at this. I am sure my team would tell you. Um, but I really try quite hard. And there's been like cases where I caught myself, uh, writing a message and mid messages going way, wait, wait, I should not be saying this.

[00:50:25] I should not be saying anything. I should just keep my mouth shut. Uh, and then I instead just set myself a reminder, uh, for like two weeks later to check back in and see how things are on. And I've found so many times that the thing that I thought, like there was a thing someone was doing and I was like, that is definitely the wrong way to do it.

[00:50:40] Uh, but I keep my mouth shut and then come back two weeks later and I'm like, huh?  That was like totally the right way to do it and potentially better than my way. 

And then the last thing might find that the thing that  You thought could have been better. They also totally think they could have been better, but maybe they have like a whole bunch of ideas on how they're gonna do it better next time. Um, I have definitely had that experience many times [00:51:00] before and it has made me go from like, oh man, uh, person X did not do such a great job on thing Y to like, huh, person X is like really, really thoughtful and totally going to do like better and better work, um, in every subsequent project.

[00:51:18] Brett Berson  are there any specific areas that you are willing to I mean, for me  the sort of, um, thought exercise comes to mind is that, you know, if you're at an offsite and the structure is burning down, you might tell someone the exit is over there. You should go there as opposed to sort of leaving it into some open ended exploration of what you should be doing, given the buildings burning down.

[00:51:45] with that as a mental model, are there things you found that you actually are very, very small areas that you tend to be more prescriptive or more willing to micromanage or it's full stop? You won't do it.

[00:51:58]  oftentimes [00:52:00] I find that the way people end up  telling others what to do is like. Telling them what to do that they already know. Um, or  they have already considered. The thing that you think is very, they have additional information, they're acting on that additional information and here you are, uh, making them either go, all right, well, you're the boss.

[00:52:21] So I will do what you say, which is terrible. Now you've lost that extra information that they have, that would've made a better decision or making them explain to you and teach you about how they have this extra information. And like making themselves defend a totally correct decision to you.

[00:52:36] Um, and I feel like that's the more common situation. So,  the way I think of it is like, if I'm going to be saying something like that, like the exit is over there, I better have more information than the other person. Right.  I better not be doing it. Uh, just to like,  leave myself feeling like I was useful today.

[00:52:55]  And this is like, especially true when you are a founder. And [00:53:00] you have generally senior people around you, um, chances are, but they're functionally more of an expert at their thing.

[00:53:06] Right? So  everyone who works for me is better at their thing than I am at their thing. And so if I'm telling  Phil, our marketer that like, oh, I hear people advertise on Google.  like Phil has heard of Google, he knows that you can pay them money to advertise there.

[00:53:23] He has almost certainly considered it. Right. Um, and that's the level of, of  micromanagement that I think is incredibly common. Um, I, I also think, not that any VCs want advice from me, but, uh, I do find VCs do this fairly often where you'll be in like in a board. Um, and the VCs will be like, have we considered, um, buying leads on social?

[00:53:46]  or like,  I think we need to, uh, close more bigger deals, right?  it's just like, oh my God, this is the least useful comment in the universe. Um, and now I have to like explain myself of  all the things we try and considered 

[00:53:58] Brett Berson so before we end [00:54:00] by talking about the talent and talent evaluation component, are there any other parts of the system that you've built, the rituals that happen on a weekly, monthly, or quarterly cadence that the way that you work, that we haven't yet talked about that

[00:54:25] Sidharth Kakkar The most important thing that we haven't talked about yet is, um, how we use feedback to kind of close the loop on a lot of things. Um, so we have a very, ritualized sort of method of feedback, which is, it happens every four weeks. Um, uh, we are about to start standardizing on like what day that happens for the whole company, cuz um, to us that's part of culture.

[00:54:48] So that's where, um, Michelle and I get to make a call. Um, and what we also template sort of what the format is. Um, it's, it's just a format that I'm sure everyone's heard of.  the [00:55:00] start, stop, continue. Um, although no one ever uses the stop, so it's really start and continue. Um, and. That feedback goes both bidirectionally.

[00:55:11] Um, so both from, from, uh, the employee to their champion and the other way around, uh, every four weeks, and this is sort of the opportunity to make up for what we gave away in the no micromanaging thing, right?  we gave up the idea that you can tell people what to do, but in this, we gain back the idea that, um, you help people improve and that thereby make better decisions in the future, um, and reflect.

[00:55:41] Um, and that means that you will never have to tell them what to do. They will, they will just get better and better.

[00:55:51] Brett Berson this feedback model?

[00:55:56] Sidharth Kakkar This is something I actually did at freckle as [00:56:00] well. Um, there actually we had a slightly different format.  we called it glows and grows. Um, so, but it ultimately ends up being the same thing. Uh, I think the, the actual words don't seem to matter so much here ended up looking pretty similar, 

[00:56:12] Um, and there, we also did it, I think we did every six weeks there, um, because we're not doing one on ones at subscript. Um, it's helpful to have it slightly more frequently. Um, and sometimes there's just like, not that much to say, and you end up repeating yourself, uh, a little bit and that's okay. Um, repetition is great for this type of thing.

[00:56:32]  the other part of this feedback model that I am a big proponent of is. I think it's easy to say. Everyone should give each other feedback, but without a dedicated forum feedback always feels a little more offensive.

[00:56:49] It's sort of like, why did you go out of your way to tell me  this thing that I'm not doing perfectly, it like feels a little offensive, but here in this template,  there's like a blank [00:57:00] space. You're supposed to put the thing that the other person can improve.

[00:57:03] It's sort of like, you know, annual reviews or whatever,  but obviously like waiting a whole year to do that type of thing, um, at a startup is not such a great idea. So, um, every four weeks turns out to be in my view.

[00:57:14] Really good.

[00:57:17] Brett Berson So I wanted to end our time by talking more about the people, part of the equation. And so maybe a place to start would be given the model that you're executing on. What is, what is required in the talent that you're bringing in. And in what ways would you say  an effective employee at subscript might be similar or different to an effective employee at freckle?

[00:57:49] Sidharth Kakkar A lot of the ways are of course similar because, you know, um, ultimately cultures often end up being reflections of their founders and was founder in both. Um, I think the, [00:58:00] the ways that are different because of specifically are a sync, more autonomous approach is, um, we need people who are more sort of writing oriented at subscript.

[00:58:13] Um, and so that's a huge part of our interview process, for example, um, in every single role, uh, we don't have a very long interview process, but one of the steps is always involving writing, um, for an engineer that would be Writing about  a pretend feature that they're working on, uh, and collaborating with a, um, you know, uh, pretend colleague on that feature, um, for a designer, um, it's a little bit of a back and forth discussion on, um, stuff that they've worked on in the past.

[00:58:43] Um, even, even for, um, sales employees and, um, you know, data team, everyone does the writing exercise as part of their interview. Um, and then the other thing on for autonomy that's really helpful is, um, having [00:59:00] ambiguous, um, parts of our interview process. So even like every step we try to make.  at least a little bit ambiguous.

[00:59:09] Um, and what we're looking to see is  how people ask questions, how people make decisions on the ambiguous parts. Um, what is  their ability to infer context based on all the other things that are talked about, um, and that helps us sort of gain confidence in their ability to make decisions on their own.

[00:59:29] Brett Berson So, so Arthur, you just mentioned this a little bit. Maybe we could zoom out and you could walk us through what the current interview process that you've developed is kind of end to end. And maybe why you've architected it in this way.

[00:59:43] Sidharth Kakkar for sure. Um, I'll give an example of our engineering interview process. There's only three steps, um, but it's, it's  pretty similar for every role. Um, the first thing that we start with is, um, we, um, set context of like how we work and you [01:00:00] know, how our interview process will go, uh, because we wanna make sure remember hiring is a segmentation exercise and  it's actually a non-trivial number of people who say, and I'm sorry, actually, this ay thing is not for me.

[01:00:11] Uh, or  this, like not an office thing is not for me, which is great. That's what you want. You wanna, um, filter out sort of the folks who are ultimately not going to actually be a fit for working in their culture. Um, from there the first step, um, becomes, uh, a technical exercise.  it's a sort of, uh, Toy project.

[01:00:32] Uh, we just took some open source project and you kind of enhance it a little bit. Um, but the critical thing is that you, um, first do a little bit of planning ahead of, uh, doing the changes to it. Um, and you communicate your planning, um, and then you make your changes. Live while on, on video. So you record yourself doing it.

[01:00:51] Um, and we kind of get to see how you think, and you talk through the decisions that you're making. Um, and we learn so much, like there's so many signals coming through on this [01:01:00] thing. The first is like, how do you think about design? Um, so not just like, you know, hacking away, but also like how do you think about sort of slightly higher level decisions, um, which is important in an autonomous environment.

[01:01:12] Um, and we find even, even  more junior candidates can have really good thoughts on this stuff. Um, the bar is slightly different for junior candidates, but we still are looking for the same types of skills and, and characteristics. Um, then when we get to watch you actually write code, um, you know, there's, there's a lot of, um, um, variance and ability there.

[01:01:34] Um, and so we get to like find people  who are actually able to like write. Code that is clean and duffle and, um, you know, aren't, aren't creating a bunch of debt or holes or  when they are, they like explicitly call it out, stuff like that. Um, so we learn a lot that way. Um, and then the communication throughout, we learn a lot about, like, how did you communicate your design decisions?

[01:01:55] How did you communicate while you were, uh, working through your decisions while you were program? [01:02:00] so that's step one. Um, and probably gives us the most signal in the process. Um, and then for step two, you take the thing that you worked on and you expand the documentation around it. Um, so here, the thing we're looking for is sort of like, um, what is the user experience of working with this person?

[01:02:18] Um, and if you, if you're like more backend focused, your first project will have been brought backend. If you're more front end, there'll be that. And you essentially work with your counterpart, uh, and your job is to create documentation that will help the person working on the other half of this feature, um, know what to do and how to do it.

[01:02:36] And so that's where a lot of ambiguity comes in as well. Um, because we're not, we're not like, you know, include these sections or whatever. We're basically just seeing like, Help the other person work with you. Um, and so you get this sense of like, how do they communicate?  how thoughtful are they, how deep do they go on certain things?

[01:02:52] How much do they anticipate your needs? Uh, as the pretend colleague, um, and this, this gives you  tons of signal of like, what is it [01:03:00] actually going to be like to work with this person? Um, how, how good a teammate are they gonna be? Um, super high signals here as well. And we've definitely had folks who  absolutely were like one of the best on step one, but one of the  worst fits on step two.

[01:03:15] Um, and  that makes a decision fairly clear. Um, and then step three is  a actually a sync chat with, uh, Michelle. And really what we're doing here is like as much trying to make sure, you know, what you're getting into,  making sure that you are in our segment, um, and 

[01:03:31]  just trying to like, absolutely make sure that you are into this. Um, and we haven't, there's, there's very few people we've hired who like it turns out were not into it. Uh, so that's really good. Um, we have hired folks in the past who turned out not to be a fit for it, even though they were into it, different story, but, um, that works pretty well too.

[01:03:52] Brett Berson With sort of all of that as,  background, what does this ultimately mean a week in the life of [01:04:00] Sadar? And what does that look like  in terms of what are filling up your days? And maybe after you explain that you can compare and contrast that with what did a week look like when you were the CEO of freckle?

[01:04:17] Sidharth Kakkar It's shockingly different actually, um, which is kind of surprising.  because, uh, EF freckle, it was a lot of meetings, um, obviously, uh, because that's how people do, you know, things and most CEOs are in a lot of meetings. Right. Um, you know, they're like joining the, go to market, like set up meeting and they're doing their all hands and maybe doing a prep meeting role in the one on ones and so on.

[01:04:38] Right. So a lot of meetings, um, and then. There was a lot more of me getting down into the nitty gritty of the things people were working on, especially in the earlier days. Um, and trying to be like, you know, helpful or more often the person pointing at the fire exit when people are already going through it.

[01:04:58] that ended up being me quite often. [01:05:00] And so, like I was spending so many cycles doing that. Um, at subscript it's quite different. I have a lot fewer meetings. I have meetings with  external folks. So I have meetings with like customers and, um, other people who I'm doing discovery with stuff like that.

[01:05:15] Uh, but there's zero internal meetings. So that generally means a typical day I have between zero and two meetings.  which is amazing the rest of the time.  a lot of the meeting stuff is replaced a bit by, um, notifications on  notion and slack and email.

[01:05:31] Um, so there's sort of the places where everyone goes for notifications, there's notion and slack specifically. Um, no matter what you do on the team, you're checking those two. And then there's a bunch of like functional, specific ones.  GitHub, um, there's, uh, Figma for designers, there's Asana for customer facing teams.

[01:05:52] Uh, there's HubSpot for sales and marketing teams,  but. CEO and as a person who was like doing each of those roles first,  I get a bunch of notifications from those [01:06:00] two. So there's a bit of like conversation, uh, where mostly I'm just doing like more context adding where people are like, oh, I don't know why this decision was made.

[01:06:07] So, you know, and I'll be like, oh yeah, check the stock out. Um, a lot of that type of stuff, uh, a lot of it is like we' to reaffirm culture where people will be like, Hey, how do we feel about, um, uh, people paying by Methodex and I'll be like, you decide. Uh, so there's a lot of like me saying, you figure it out.

[01:06:23] Um, and not letting people. Tell me to tell them what to do. Um, so there's a bunch of that. Um, and then there's a lot of deep thinking time, which is the greatest. So I do a lot of like, uh, writing either for myself as I'm processing or for the team as I have getting closer to a thing, which I think, um, would be really helpful for a lot of people to give input on so that a four mentioned journal of product market fit.

[01:06:48] Um, I end up writing a lot of entries in there. Uh, I end up writing a lot of like thoughts in, you know, a bunch of different areas. Um, and mostly like my number one job is to find amazing [01:07:00] product market fit. So  I think about like, what will help us get there? I do a lot of thinking about our goals about our weekly planning.

[01:07:06]  those are the highest leverage activities I do. Right. Like thinking hard about our goals and, um, aligning everyone on those. Um, when we're hiring,  Often, if it's like someone who's gonna report to me, I've done a, spent a lot of time doing that. Um, but when we're not hiring, 

[01:07:20]  then I will also end up doing a little bit of like, um, stuff that  I just enjoy doing. So I really love programming. Um, so I just go pick up a ticket and work on it. Um, and, uh, that has been, uh, quite delightful as well.

[01:07:36] Brett Berson Maybe to end, as you think about this model, this culture, this way of behaving and operating that you're working on at subscript, what are the, what are the things you're still wrestling with? Or don't don't know the right way to do something is.[01:08:00] 

[01:08:10] Sidharth Kakkar  a couple of things that, um, I think we talked a bit about that I haven't quite figured out yet is how do you apply, uh, our beliefs on autonomy and not micromanaging to more junior members?  I think the principles will still apply, but the specifics may vary a little bit more. Uh, so there's a bit of thinking to be done around that.

[01:08:31] Um, and figuring out to be done around that. Um, there's, you know, we talked a bit about, like, there are no sort of random serendipitous encounters in an, uh, in a remote world. 

[01:08:42]  we're also thinking about sort of what is the right level of, um, in person together time? Uh, especially when it's so hard in a COVID world, um, for people to travel. Um, so we're still thinking about that and  mine and Michelle's inclination is [01:09:00] for more, but there's like real sort of, um, operational limitations,  that we have to sort of figure out.

[01:09:07]  and then there's like, um, stuff like, um, Taking cultural things that, um, Michelle and I are very good at because we develop the culture and making sure that that extends to the whole team, because we want almost everything to be, you know, very autonomous and people making their own decisions  but like some things are really hard to provide, uh, contextual guidance on.

[01:09:34] Um, for example, the hiring bar is a really hard one, right? Like, um, in lots of big successful companies, there's places where CEOs famously, like, I don't know, interviewed every employee until a thousand people. Like, I don't think that's the right approach, but I understand the problem they were trying to solve, which is that this is a, it's a hard thing, um, to like, um, SUSE your culture with.

[01:09:56] And so you have to find good ways to do that. And we are thinking about that as.[01:10:00] 

[01:10:01] Brett Berson Great place to end. Thank you so much for spending all this time with us. This was one of my favorite conversations. This was really, really great.

[01:10:08] Sidharth Kakkar thanks, Brett. Thanks for the great questions.