The biggest lessons from building Hubspot, from co-founder harmony to engineering the culture — Dharmesh Shah

Brett Berson: Well, thank you so much for joining us. I can't wait to dive in here.

Dharmesh Shah: Thanks for having me 

Brett Berson: I thought, uh, an interesting place to, start our conversation is to talk about your co-founder relationship with Brian and sort of what I've noticed that at afar, it seems like an outrageously productive relationship over an exceptionally long period of time. even when you have quite productive, uh, founder relationships, you often don't see them, go on for 15 years.

And so I thought it might be interesting to have you talk a little bit about when you look back on what made it work and what made it so, uh, productive, what sort of the root cause? What, what made it work?

Dharmesh Shah: yeah. you're right. It has been a long enduring, Um, productive relationship. And I think that we'll start with the, kind of the obvious things which people don't talk about, but apply one is when you're picking a co-founder or trying to choose who you're going to partner up with, on this kind of long startup journey, you need to actually like the person, and you need to have some degree of mutual admiration and respect.

And I think that gets missed sometimes when founders are trying to kind of assess each other, they'll assess the idea. They'll maybe have a beer together. and that's all well and good, but you have to kind of really ask yourself, I enjoy spending time with this person because you will be spending a lot of time with this person, in many cases more with that individual or a group of individuals than anyone in your life, including your significant other.

And so, you know, just do you enjoy spending time? That's a fundamental thing. and the other thing that I think is interesting in terms of figuring that out, I don't think people allow themselves. enough Time to get to know their, potential co-founders. Um, and Brian's my co-founder Ryan Allegan.

He and I were very, uh, I'll say methodical about making this determination. So, uh, we met in grad school, uh, which I think a lot of founding relationships start there, but the reason that academic environment is useful, um, when used correctly is it gives you a relatively lightweight non-intrusive way to kind of really get to know a person, at least in a, even though it is a somewhat, uh, hypothetical setting.

Um, but what we did is that we said, okay, well, we might start a company together someday. Uh, we knew I still had another year left before I graduated, but we said, okay, well, every class that we're having together, which was a lot of them, um, any class that has a project, we're going to intentionally put ourselves on the same team.

And sometimes it was just a team of the two of us. Um, and we took it a step further. We said, well, every project that's kind of required of us, we're going to. Make that project tied back to this startup that we called HubSpot, the HubSpot wasn't a company at the time, it was a idea for a company. Um, and then that idea shifted.

But the idea here is to try and, uh, I'll use the word simulate for lack of a better term. It's like, what would it be like being around this person for 4, 6, 8 hours at a time? What would it be like having a project that you're working On together that's going to be measured and that you have to deliver that actually does matter.

Um, what would it be like and how do they change as circumstances change? Um, and it's a very compressed one or two year environment that you get this opportunity, but we maximize that opportunity and, and that's, um, I think that's what contributed to it. We sort of knew each other really well. And in our personal relationships, we often will, you know, date for a long time before we make a very long, um, ideally, you know, lifelong commitment, um, in, in, in co-founder land.

I don't think people are as deliberate about, um, spending time with their potential co-founders and, and getting to know them. That would be it it's like, make sure you like them, uh, make sure you have yeah. Mutual admiration or respect. And then on a tactical level, make sure you have kind of different skills or divergent skills, uh, but common values.

And this is going to sound a little bit cliched because it is, but this kind of different skill sets is important. And that's an easy one that says, okay, well you both, both shouldn't have the same set of skills, the same background. You can't be clones of the same people. Um, it's often tempting to do that with co-founders because those are the kinds of people you gravitate to.

And you know, when you hang out with, um, in our case, you know, Brian grew up in, in sales and marketing. I grew up in product and engineering. And so we had very kind of different backgrounds, um, and in different skillsets. But then we also had a kind of common set of values, which is super useful. Um, as you're scaling, you know, one of them being around transparency, one of them being around this.

Obsession with the customer and solving actual problems. Um, so it's all those things kind of help. You have to know your founder like them and hopefully be different enough that you can bring different things to the table.

Brett Berson: On that point about common values. Did you explicitly surface them in that year or two before you started the company and they kind of express themselves organically it's a good question. So a couple of them, like this notion around transparency was there from time T equals zero. And it was not because we sat down and said, oh, here's our list of values. Transparency is one of them. It was more, a matter of convenience, which is, Hey, uh, we sat down and we had to make a decision when we started the company.

Dharmesh Shah: It's like, okay, well we have this information. We just hired our first time. And we have to decide what we're going to share with this person. Now we could have gone one of two ways, right? Just a I'm an engineer. So I kind of look at the boundary conditions. One could have been that we're not going to share any information with anybody.

It's just the two of us. And that's just how we'll run the company. And, um, you know, it can be argued whether it's actually possible to run a company that way, but we'll call it the minimum amount of information, uh, to all possible people. The other end of the extreme was to share all possible information with all possible people.

Um, and the way we kind of looked at it, the, the latter was a much more efficient way to go about it, which was, we didn't have to make the choice. We didn't think the first one was going to work in terms of being closed. We wouldn't be able to delegate. We wouldn't be able to scale. Um, and so we kind of tested.

It's like, okay, well, how would life look if we said we're going to make everything available to all people, unless it's illegal, um, or

it's not our information to share. And so That core value just came out of just a practical, how do we want to run the company? What does this feel like? And then that translated over time and to. Kind of an articulated, here are a set of values and transparency has been one of them since time, time equals zero for HubSpot.

Brett Berson: That you too could have been a successful co-founder pairing working on most types of companies or was this specific company in HubSpot, a unique expression of kind of the way in which you two, um, could be successful.

 So in theory, uh, Brian and I both believe that we could have been parachuted into any kind of similar tech company. That's our, both of our background, um, and would have had a reasonable shot at success will still be the idea itself. And the other connectional factors were, were common.

Dharmesh Shah: So the answer is yes, I believe it that, uh, we could have been at a different company. It didn't have to be this one on a practical matter. I have kind of a semi evidence of this. What do we started HubSpot my kind of motivation. So just taking a step back, you know, I had done a couple of startups before, uh, sold them and I said, I'm going back to grad school and to kind of find myself and figure out what's next.

I had promised my wife, I would not do any more startups. That was the agreement. Uh, she and I have been together since, before I did my first startup. Right. And I'm like, no, I'm not. That's, you know, I live the, the kind of proverbial, canonical startup life with all the startup hours and all the sacrifices people, um, talk about.

And so I said, I'm not gonna do a startup. And here I am in grad school. And then, you know, just a few months in, you know, Brian, I had met and we had this kind of a set of common passions around, you know, startups and tech and small business. Um, and so we first made the decision. This is why this kind of deliberate kind of feeling each other out and getting to know each other, um, through an academic setting and these various projects, but we didn't really know what the idea for the company was going to be.

Right. So. In a way you can think of it. Like Brian, I first came together as co-founders and then jointly came up with, okay, well, we've decided that we really like each other enough and that this is kind of worth pursuing. We have broad sense of what we wanted to do, but the idea that's now HubSpot didn't really manifest until after we made the decision that the two of us were going to start a company together. 

Brett Berson: I'm interested when you think about your. Kind of compatibility and your worldviews, how did you approach if you did this and maybe it was completely unintentional the, the path of growing together over a long period of time. Um, and I, I think in some ways, as you were sort of mentioning, it's, it's similar to a lot of relationships and, and potentially marriages, and we all hear about marriages where people are super aligned when they get married, but in 10 years they're different people.

And I think, you know, you had a company that went from zero to one to 10, to one of the most meaningful, um, SAS companies in the last 20 years. And so just an insane amount of change and insane amount of scale. And yet it seems like you were able to productively grow together. And I'm curious, sort of any reflections on that.

Dharmesh Shah: Yeah, I think a few things contributed. One of the things that.

brought us together in the first place is that, um, and, and it's not overly surprising since we were both in grad school when we met, but we have this, um, real passion for learning. Um, we're kind of learning machines and then it's, uh, and, and the nice thing about as you know, the company scaled is that, um, there was just, and this was our first time doing something, you know, anything approaching the scale of what HubSpot became.

And so, you know, every, every month, every quarter, every year there were new things to learn things that neither one of us knew, or in some cases. Um, it was an area that, you know, Brian had more experience in, or I had more experience than the other one kinda came along. Right. So, um, we kind of both learn from each other and then both learn together about things that neither one of us knew much about.

Um, and so that's been helpful. I think that, um, kind of have a sh having a shared purpose and it's not such like a highfalutin kind of mission, which we did have, uh, but we had this kind of shared goal of trying to get better. Um, and, uh, you know, to this day we will send each other emails at one or 2:00 AM.

We'll send each other book recommendations. We know what's on each other's reading lists. Uh, we'll uh, discuss, you know, talks we've watched or things we've learned. And we, um, and then this carried it. Wasn't just a, the co-founding team. All the early employees, we had this very kind of academic oriented.

Uh, culture. So we really believed in debate. We really believed in, um, you know, the easy issues were kinda easy to knock out. It's the ones that were where we couldn't decide on, um, where it, you know, debate was actually very healthy and it was very respectful, but it was a very academic debate. And we could pick anyone off the management team at the time and say, okay, well, we've got this issue, uh, argue whichever side, like in, uh, they may not necessarily agree with that particular side, but just like debate, um, debate school, you learn to do that.

And so that process of just learning together of kind of pushing on each other, um, even when we disagreed, which did happen, but not all that often, uh, I think that helped kind of keep the relationship productive, keep it fun. Um, because we both enjoyed learning and then, uh, growth itself is kind of funny.

It was, uh, you know, if the company had gone really, really poorly, I think we would still have survived. It probably wouldn't have been quite as much fun, but, um, yeah, that's, that's it it's just learning together is, is a useful, a useful. 

Brett Berson: can you share a story that kind of brings that to life over the past handful of years?

Dharmesh Shah: Yeah. So it's, you know, the longest running debate in HubSpot history, um, was in the early years?

and it was a deciding on our target market and we had it, uh, kind of reduced down to two things. One, there was a third one that we all had read. Wasn't going to be the target market, which was these kind of developer it folks at the time we said, okay, we're going to focus either on professional marketers, because we had a marketing suite of applications that was our first, uh, kind of set of products out of the gate. so we could, uh, focus on professional marketers or we could focus on kind of business owners and entrepreneurs. Those were the two, um, two target markets and we had names for those personas. So we called the marketing persona. We called her marketing Mary, and we call the, uh, kind of the business owner, owner Ollie.

So maybe he owned a small accounting firm or something like that, as an example. And. There were trade-offs on both sides, right? There were many more owner Ali's out there than there were no marketing marries because there's lots of, kind of startups and small businesses, but marketing, Mary understood the product better.

She had a more kind of burning need because it's like, oh, I'm trying to do marketing the old, stuff's not working. Let me look at HubSpot. Um, and so that one was one we, um, could and did literally debate for like years before we made a decision. Um, and then this was that, that, that long running debate was turned into a, one of the most popular Harvard business school cases, which has taught to every HBS grad that graduates now as part of the core curriculum, uh, in the marketing class there.

Um, but, and the way we kind of ultimately resolved it, like step one was we had to decide to decide, because previously we had kind of walked into rooms like, okay, we're not going to leave this room until we have a decision. And we like spend seven hours talking about it, uh, order food in and, and we kind of leave.

And that night. We'd be pinging you back each other. It's like, I know we made this particular call, but you know, what about this? And what about that? Um, but we finally did make the decision. Okay. Like sitting on the fence is, is just not a great way to do this. Um, and we made, made a decision and, uh, and, and moved on, but that's yeah, that's one example we had, we have those kind of long running debates.

Uh, we're much better about it now in terms of getting to a decision faster and we're much, much better about not, um, kind of second guessing ourselves. And once the, we use the phrase kind of sailing the ship, once we've failed a sale, they ship on a decision. We don't call it back to back to Harbor.

Brett Berson: so you're touching on a few maybe rituals or practices that you and Brian sort of shared over the years. I'm curious. Maybe you could expand a little on that in terms of the things that you did on a weekly, monthly, quarterly, annual basis that provided fantastic dividends for the.

Dharmesh Shah: Yeah. The, the biggest one where know we are both, um, make people night owls. Um, so that kind of helped. I'm not sure the.

relationship would have worked as well. Probably we would have still made it work. Um, so that was, that was one thing. So we'd have these kinds of late night conversations and they were, when I say conversations, we almost never talked on the phone.

Um, cause neither one of us are our phone people. Um, and that's one of my quirks is actually do like almost zero, like one-to-one calls. I never answer my phone. I have made. Three four calls a year. Um, it's a little bit of weird, but anyway, so all of our conversations were asynchronous and this is in the kind of pre slack days.

Um, so it was all over email. And so we would have these kind of, and we were never apologetic about long emails. It's because that's sort of how we thought about things. So someone might kick off a particular locale. What do we think about this particular kind of market director or this new competitor that's come along?

You know, and it could be at the 50,000 foot level, uh, talking about kind of long-term strategy and it could be down in the weeds about a particular feature within a particular product. Um, but what was very productive is the fact that, uh, it was an asynchronous. So tactically that just works better rather than trying to schedule time.

Although it's a lower bandwidth conversation, um, you know, it takes longer, but the upside to it and we go back, um, through the kind of animals. Got HubSpot history all the time. Like we'll go back to old emails that, um, especially as new folks join, because there were things that kinda, we had uncovered in earlier conversations that Brian and I have had.

Uh, and the nice thing about that particular kind of tactic of using email as a kind of recurrent, um, collaboration mechanism is that we can then take snippets, um, or even entire email threads and pull a new person in. It's like, okay, well, Brian had been back and forth on this thing around, uh, some sort of product strategy thing, and let's bring in, you know, X or Y from the team, um, kind of catch them up or B get their take.

And it's a nice, smooth way to make that process happen. So that's worked out really well for us in terms of having that open collaboration kind of, and kind of putting our brains on speakerphone in a way by virtue of kind of writing down our thoughts in an email. So that's the one thing that jumps to mind, but then just books.

We read a lot. We both have always enjoyed reading prior to prior to HubSpot. And so, and I think this came a little bit because, um, we were such recent grad students. One of the values you get of having a co-founders and by the way, so we were both at MIT Sloan, um, and this was both a good thing. And a bad thing is that the next six hires that we had, or six out of seven hires in the early team were all MBA.

Students all went to MIT Sloan within like a three year, four year time horizon. We weren't all there together. And there are lots of downsides to that in terms of that was too much too homogeneous a team. If I had to do it all over again, I wouldn't do it that way. But one upside to that model is that we have.

Uh, a higher order vocabulary, having taken the same classes with the same professors learned the same business cases, read the same books, um, as part of the curriculum. Uh, so we could talk about, you know, something in like a sentence that would carry this kind of richness and all this kind of context with it simply because like, we had just done that case.

It's like, oh, this is like the capital of one case where, you know, we're, we've got an arbitrage opportunity so we can provide, you know, they're, they'd provide like the most efficient interest rates to make things fair for consumers. We could do that over here. And just by saying, it's the capital of one case kind of carried all that kind of richness of context with it.

And that that was useful. And so we took that idea of like, oh, it was very useful to have taken the same classes and read the same business cases and things into books. I was like, okay, if we read the same books, we can kind of point back and reference either entire books or sections of books or quotes from books and say, okay, well, this sorta applies to us.

What do we think? Um, and we've carried that not just between the two of us now, the management team at home. Uh, does that on a very regular basis, we don't do a like required reading, but we do highly recommended readings. That's like a, if you really want to be kind of part of the Richard conversation, it's going to be helpful for you to have read X, Y, Z book.

Brett Berson: one of the other things that I think you all have developed over the years is, is an approach to your own annual review. Um, and I was interested in maybe if you could talk a little bit about what does that practice look like and what are the pieces

Dharmesh Shah: Yeah. The hard thing about annual reviews, uh, for, for founders and, you know, people higher up in the organization is that like you, you don't have a boss per se, I guess you could get your board. If you have a board to write a review for you that that has its own set of challenges with it. So what we did, uh, and have been doing, uh, between Brian and I is that we write each other's reviews.

Um, and we spend a lot of time on it, uh, because one of the other values we share is we both really, really believe, uh, And crave feedback, uh, when we like to iterate. So we'd like to kind of know how things are going. We love, um, love feedback. And so it's hard to get that feedback. Um, you know, when you're, when you're a CEO, you're a, co-founder just, um, from an org structure perspective.

So what we decided to do is that we would do each other's reviews on an annual basis. Uh, but it wouldn't just be, oh, here are my thoughts on Brian or Brian's thoughts on me, that person, whoever was doing the review was then responsible for talking to 15, 20 people in the team of their choosing, who they thought might give kind of most, um, color and context to how things were going.

Um, and then it was a of review writer's responsibility to kind of synthesize all the feedback, heard, add their own things in there, and then write this 10, 15 page review that we shared with each other on an annual basis. And then, uh, and then it would be a response to the review, uh, in written form that says, okay, well, I heard what was said.

Um, here are things tactically that I'm going to try and do over the course of the next year, um, to try and kind of improve those things. And there'll be, and, uh, this is a somewhat funny, funny thing is that, um, so when I write Brian's review, since I'm a product engineering guy, I write as a, as a kind of a product report and it's like a bug report almost.

It's like, oh, here are all the kind of bugs, uh, that people have reported back, right. About, uh, you know, the way you lead or the way you manage or the way you are. And, uh, just like if it were, um, uh, a kind of a product feedback thing, you know, some of those things Brian will come back with and rightfully so, he's like, well, that works as designed.

That's, um, I'm unlikely to change that. I do the same thing in the, in the reverse. It's not, not picking on him, but, um, and then we, and so we would version it. It's like, okay, here's Brian 4.0, and here are the issues. These are things people love about Brian 4.0, this new feature that you added over the course of the last year or two years, it has been received very well, but this. W we think as a bug, um, multiple people have cited the same bug. I think it's kind of holding, uh, holding you back and holding the company back. And, uh, one of the many things I love about Brian and he's, he is exceptionally good at receiving, uh, receiving that feedback. But, uh, but so the short story is, um, we spend the time, we understand how important it is.

Um, and there's some things that are just really hard to completely delegate to like a head of HR or, you know, someone else. And this is a practice that has worked well for us. And it also serves to kind of keep us closer and we can, uh, because we've been kind of friends and have worked with each other for so long, we sorta know, uh, where that feedback is coming from.

We know it's coming from a good place.

Brett Berson: when you're, when you're collecting the feedback from other folks in the organization, what are you asking them? What have you found gets you kind of the most crystallized insights or the, the most accurate bug reports from, from other folks in the organism?

Dharmesh Shah: Yeah, it's funny. So in the same way I described it to you, I, uh, this is how I describe it to the people that I'm getting feedback from. Um, especially folks that are newer to the organization, which is, you know, I'm writing a, and I'll joke about it. Like, you know, I'm a product engineering person, so I'm writing a kind of product review of, of Brian and, um, and you know, what do you think is working?

What are the features of Brian that you kind of really enjoy that it's like, you're glad they're there. What are the things that are annoyances that, you know, if you'd like to fix, but they're not really keeping you from kind of doing your work. And, uh, and they're not critical, you know, we can kind of rank the severity of them, but I try and leave it.

Open-ended um, what I'm trying to do is not say, okay, well, I want to collect feedback from this particular person around how Brian interacts with the, the engineering team. Uh, even if I'm interviewing an engineer, um, I think. That kind of pigeonholes it a little bit too much. You shouldn't kind of request certain types of feedback from certain roles, because a lot of times we think we know what the interaction between, um, you know, Brian and let's say person X would be because of the role of the organization.

And it turns out, you know, some times people have much more useful feedback that might be outside of their immediate role or their immediate set of interactions. Right? Sometimes the feedback is not because of direct interactions they've had with Brian, but it might be around, they've heard Brian speak at public events or they ran into him, um, at a conference or whatever it might be.

Uh, so I try to leave it as open-ended, as I possibly can. Uh, it makes it a little bit harder to synthesize because it comes back in a relatively unstructured way. Um, that's part of the fun of it. And then you kind of look for patterns that says, okay, well this was unaided feedback like that. I didn't try to lead the witness in any way.

Um, you know, what came back here, people here are the things that kind of jumped to people's minds and they tend to kind of self-select in terms of things that are most impactful versus trying to answer your question. Uh, so I try not to be too specific. 

Brett Berson: When you think back to all the bug reports that you've received over the past 15 years, are there specific bugs that have led to the most significant growth for you personally?

Dharmesh Shah: Uh, yes. And in both ways. So I'll, I'll give you one of each one is a reported bug. Um, and the bug was, we don't see enough of Dharmesh, uh, when he's in the office. It's awesome because we feel like we collaborate. We feel like we're connected. Um, and we really enjoy, and this is pre pandemic. Um, and I got that feedback for at least a couple of years, uh, and we'll keep popping its head up every now and then.

And my response to it has been that sort of works as designed. And so I understand the value of that kind of in-person meeting. But, um, the cost of that particular, uh, fixing that particular bug is exceptionally high for me emotionally. And as a result of which is exp exceptionally high for HubSpot, because that kind of one or two hours of in-person time on a one-on-one meeting or I'll do it.

Uh, but it's going to kind of ruin the rest of my day. Maybe the, maybe my week, because I am a, I have a lot of quirks, but one of them is around, um, I'm a super introvert like it's, and, and it's not that I don't like people. Um, it's just that I don't like being around them a lot, um, in, in, in person. So I'm great, you know, over long email exchanges, uh, I can, you know, hop on the slacks every now and then if I need to, but just that, um, it just takes a lot of energy out of me.

So that was one that was like, okay, I understand that this is what you want. But, uh, my apologies, I, I don't think I'm going to, I'm going to deliberately decide not to fix that particular one. Um, one that I did fix, which is around like Dharmesh, you know, you're kind of out there. We really don't know what you're working on or what you do or what you care about or like it's.

Um, and that was that because I was kind of hiding anything. It's just because I wasn't sure anyone really cared. I didn't want to distract the team with whatever random thoughts were kind of going through my head. Um, and so now what I did to fix that particular bug and still kind of aligned with my original, you know, like my alone time introvertedness is I started a series of Wiki posts, and then I write on the internal, um, it's basically an internal blog.

That's called Dharmesh as ponderings. And I'll talk about, um, things that I can't actually talk about publicly because it's, and, but the nice, good newsletters inside the company, um, doesn't leave the company and that's been a great way for me to be transparent. Um, and then people can opt in if they want to read that blog post on what I think about web three or what I think about whatever it happens to be on my mind at the time where I like, you know, made a particular decision I made, um, they can kind of get inside my head if they choose to.

And, uh, and that's, that's what.

Brett Berson: when you think about your own personal growth and evolution, when you get bug reports that you decide are worthy of you acting on. Do you immediately go to some system level change or practice in, in sort of like the example that you just gave us around, um, more internal comms to, to keep people abreast of what you're thinking about or what you're doing, or are there other ways that you actually go about implementing feedback?

And I'm always curious about this because I think in general, even if folks are good at receiving feedback, most people don't actually make a profound change and they just, you know, if they're not detail oriented, they kind of keep going on, not detail oriented and kind of implementing feedback is, is maybe trickier than one would imagine.

Dharmesh Shah: Yeah, I think this is, uh, you know, we have some kind of base level features. What I call the HubSpot operating system, which is our culture and how things work at HubSpot. Um, and one of them is around, you know, we believe in autonomy, um, which is super important. So people have lots of discretion and control and lots of companies do that.

But we think the flip side of, um, autonomy is accountability, which is, um, you have to kind of hold yourself and your peers and your leadership kind of accountable for the things that they commit to doing, um, and accountable to our values. So kind of my motivation for doing that kind of series of Wiki posts around transparency.

It's like, okay, well, you know, transparency, wasn't hard. It's like, okay, well that's our culture. So I don't think anyone's gonna be bothered by the fact that I'm putting my kind of thoughts out there. Um, but in, in terms of the review responses, one of the things that Brian and I have done is, um, when we decide we're going to make some base level changes or like we're going to respond to, um, you know, uh, feedback from the review, we will share that, um, with all the employees, right.

We'll say, okay, well, here's what I heard back from my latest review from. Here are the bugs that are, you know, you might call them bugs, but I say they work as designed. Um, I'm not gonna address them and, and here are the things I am willing to fix. Um, and often I'll try to put a timeline around it if I can, if it makes sense, but most it's assumed that that'll be in the kind of following calendar year.

Uh, but it's, that's what I think, make sure that the change that should be happening is happening is that it's okay for someone not to make the change and they can kind of make their case for why they're not going to change a certain thing, but they actually have to kind of reveal that they made that decision.

Right. You can't just ignore things. Um, and so, and we do this, not just at the individual, you know, Brian and I kind of review level. we do a, a net promoter score survey for all of our employees. Uh, we do that poorly and have been doing it for over a decade now. And. That's basically like a bug report, um, for, for the company right there, just like you would have, um, a customer satisfaction survey or a net promoter score for your, for your product.

Uh, we do the same thing and we think of culture as a product and people submit their feedback and we will, we, as a, as a management team will say, and we publish it. Uh, the, the responses themselves are anonymous, but the actual entirety of that survey result is, is published both in quantification format of the classified form and say, here are the things we heard back, and here are things that absolutely, uh, many of you felt this way.

It's a big issue. It's a hard one to fix. And so here's the plan we're putting in place to, um, to make those changes. And then similar to how we do it on the one-on-one feedback. We'll say these are things, a handful of, you mentioned we understand where you're coming from. Uh, but we just don't think we're going to be able to make that particular change.

At least not. Now, we're not saying not ever, uh, but we're not going to be able to get to it. Um, you know, anytime, anytime soon, let's talk about it again. If it's still continues to be an issue. time. 

Brett Berson: before we switch over We didn't touch on your own role as a co-founder in the business and, and maybe how it changed over the last 15 years, or maybe in what ways?

Hasn't it.

Dharmesh Shah: Yeah, this is, it's a good tie back to some of the co-founder decisions as well. And then the relationship. So when Brian, I decided that we were going to start HubSpot, make it official, and we made a couple of decisions. One was, uh, we're not going to start the company. I was still in school. He graduated a year ahead of me.

I still had a, uh, a thesis. And so, and since I had promised my wife, I wasn't gonna do a startup. And then I kinda convinced her. I'm like, okay, I have one more step up to bat left. Um, and I've met this great guy. And she had met Brian several times. Uh, but once we made the decision to kind of start it, we said, okay, well, I'm not going to officially start it.

until literally the day of my graduation, which Was June 9th, 2006.

Um, but then we had this kind of, which I think all co-founding teams should have is to sit down and to have this very deep, um, discussion and go over like the common questions. It's, it's amazing to me, how many co-founders have not talked through some fundamental questions. One of which is like, okay, like, what do we want out of this startup?

What does success look like? Uh, what happens if a year from now someone comes over and offers us a hundred million dollars to acquire the company? What happens if three years from now, one of us is just kind of disengaged with the company and really doesn't want to do this anymore. What happens if we fundamentally disagree on something.

Like a super pivotal, like all these questions that are, um, you know, super important and people are like, ah, well, you know, we'll cross that bridge when we get there you can certainly do that, but it's actually much, much better to get those issues out on the table, um, as early in the process as you possibly can.

Um, because it just gets harder later. Um back to your question around my role. So one of the things we talked about, um, in that kind of early set of founder meetings was, uh, first well who's going to be CEO, and this was an easy one because Brian wanted to be CEO. He had never been CEO of a company before.

Uh, and I did not want to be CEO because I had been a CEO before, um, of my prior two startups and I suck at it. That's just not my thing. Um, and so we then took that a step further and I said, Brian, you know what? And then this kind of came to me in the meeting. Um, I don't think I want any direct reports.

It's like, I don't want to manage anyone. And my motivation for doing that was like, okay, well, you know, I had run, you know, my prior startup for 10 plus years as founder CEO. And one thing I've learned about myself is I'm really bad at management and in the kind of classic kind of textbook way. And the conversation I had had in my head was, you know, I could problem.

I'm a reasonably smart person. If I spent 3, 5, 7, 10 years, I could probably get okay at management with some coaching and some training and I could become passively okay. At it. Um, and I didn't want to spend five years and a bunch of calories getting passively okay. At something. And, uh, so Brian, I had this conversation.

I'm like, Brian, why not? I just don't manage people. I'll be an individual contributor. Um, and I'll be a hundred percent in the business. Um, I'm not trying to shirk any work or try to have any kind of side hobbies or anything like that Um, and so that's the decision that we made. So. HubSpot's 15 plus years old.

Now we have 5,000 employees and exactly zero report to me. And if I had to trace back, you know, one of the reasons, you know, the kind of relationship between Brian has worked, um, is that early decision, um, is that, you know, I kind of decided to focus on my strengths. I decided to not try and work on mitigating that particular weakness.

Um, and had I not made that particular choice had HubSpot not made that call, um, uh, collectively I don't know that it would have lasted that long. That must start up. you kind of guy. I think my life would've gotten increasingly more painful in ways that I don't like. Um, and I probably wouldn't have lasted as a startup guy.

So, and now, candidly, I'm having a better time at HubSpot now, even though we're 15 years and, you know, 5,000 employees that I was having in the early years of, of HubSpot, And I even, um, Brian, I talked about that particular kind of observation, which is, we're both having, uh, a lot of fun and these got later years it's because, well, you know, those early startup years can be fun and yes, you're a super nimble and that's fine, but there's some value to having scale there and being at the grownups table where we can say, oh, we've got this idea.

We've got this thesis that have the resources and not just like money, but having people and talent that we can put against a problem and put against an idea that we couldn't do as a 10 person scrappy startup. There's only so much you can do. Um, but now we can actually put big bets down. That's awfully fun.

Brett Berson: Was there anything that you did, um, that created a dynamic where you, as the co-founder could be. Effectively and individual contributor throughout the life of the company. And I think that, you know, so many norms that companies, uh, ties influence with size of org chart that you are responsible for. 

Dharmesh Shah: Yup. 

Brett Berson: And so I would think there might be a little bit of just kind of peculiar dynamics in the first few years, or maybe even at scale. Given how it's quite non-traditional, I'd actually say in working across hundreds of companies, you tend to see the co-founder and CTO effectively take on the BP role early on and struggle with it mightily.

And it leads to some fracturing or implosion, um, where that person ends up leaving the company. And so is there anything that worked for you that made it really effective to, to kind of boldly take on this? I'm going to contribute in this specific way?

Dharmesh Shah: Yeah, and I think it, um, this is going to sound paradoxical. Um, it, it starts with a degree of, I guess, a self-awareness, um, I accepted the fact that I'm not going to be a great manager and then they kind of, the next step was like, okay, what would life be like, um, as an individual contributor and kind of sticking to it, it's like, okay, I made that choice and 

I think part of the thing that we have to kind of keep this in mind is that, uh, founders have a special kind of designation within, uh, within a company just by virtue of having been founders. Um, and so they, their kind of opinion, uh, regardless of whether they have managerial responsibility or not, it's going to carry some weights, right.

Both, um, while they have operating roles and even post operating roles, if they're just on the border and exec chair role or something like that, um, they're going to be influential. They're going to have, um, so I'm not sure we would be able to pull that off as readily with any, any other, even senior exact, let's say we had a, a brilliant, um, you know, senior person on the engineering staff and it's like, okay, I'm just going to move to a pure individual contributor role.

Um, and, and, and they can certainly do that. Uh, but then the impact may not be what they want in terms of what the influence on decisions and things like that. So I think that the founders have a somewhat special designation, but, but we have kind of embraced this idea of. Um, especially within our engineering org is that you don't have to go into management in order to progress in the company.

Right. We recognize that, you know, your progression and the value you get from the company should be a function of the value you're adding to the company. And then sometimes people are great managers. They should do that. Sometimes people are great at building teams, they should do that. But then some people are just really, really good at building great products and they should do that.

Um, we shouldn't force them into something simply because the conventional wisdom says, you know, your value is a function of, you know, how many people you have downstream and the org chart. Um, we just don't think that makes sense. 

Brett Berson: Before we close this first chapter, the conversation around co-founders and early people stuff. Is there anything we didn't touch on as it relates to advice you would have for folks consider looking for co-founders figuring out if someone should be their co-founder having a productive relationship.

You mentioned just briefly a few minutes ago, the idea of these fundamental questions that you tend to think should be discussed before you start a company versus after maybe we can talk a little bit more about that or, or other things I'm sure you get pinged all the time and you probably write long emails back, uh, when people are asking you about this to avoid a phone call.

Um, and so in those emails, is there anything maybe we didn't explore that that might be useful to share.

Dharmesh Shah: yeah, a couple of things. One is we'll talk about the kind of negative cases. I think where things tend to go wrong is when you talk someone into doing a start-up with you, um, and you know, I'll, co-founding relationship involves some element of sales. You're talking each other into doing something, but if it's too one-sided, which is okay, you're super passionate about this idea you happen to have come across this person that you think is a brilliant engineer.

Brilliant. Whatever they happen to be, um, that kind of love needs to be reciprocated. Um, it's okay. If they're not as kind of gung ho about doing a startup or about your particular kind of idea or mission, uh, but there has to be, it has to be like more than a zero or one, right? Otherwise you're just kind of pulling them along and that won't last, you may talk them into it because you're a charming individual.

Uh, but there needs to be this kind of fundamental desire that they have, and they need to see the upside of, regardless of where the startup winds up going. And this is the thing I use in kind of recruiting people, um, at HubSpot, even, uh, even at scale, which is, you know, we've like many companies, we, we love entrepreneurial folks.

We love the creativity that comes with folks that are super entrepreneurial. Well, if you bring on those folks, it's like, you're going to have to let them kind of learn. That's the thing they're solving for and be okay with the. Yeah, things might not work out. So the people that you want on board are the ones that even if the startup doesn't go well, which, uh, many don't, um, for any number of reasons that they won't have regretted it, right.

They may regret certain decisions, but not the one to actually join you and start the company that they will have learned enough. And their personality is such that they will have extracted enough value and gotten enough from it that regardless of what the outcome is, they won't resent or regret having made that choice.

Um, and if you kind of sold too hard to try and convince them, um, that often becomes kind of counterproductive as, as you go through the trials and tribulations and the, you know, the rollercoaster ride that is, uh, that it start-ups. So that's one, it's just, don't sell too hard, make sure they want to do it too.

Uh, that's thing, number one, uh, thing, number two, I would suggest is you need to find a way to date before getting married, um, is that, and you know, maybe you don't have the luxury of being in an academic environment for, you know, a couple of years to. But even if you simulate it somehow that says, oh, look we'll, and I'm making things up here, but take a, take an online class together or work on a project together, something smaller.

Doesn't have to be, you know, this startup, um, and maybe even collaborate on something, uh, that helps you kinda understand what makes them tick. And a lot of those questions that I talk about, you know, co-founders having with each other, I think prospective co-founders should probably have with each other, like even earlier, before you even decide, it's like, okay, well, you know, like, what do you think about that?

Like if, if someone came up and offered us a hundred million dollars, like what, you know, how does that feel? And I would much rather kind of filter out the folks where you're not aligned. There are no right answers to some of those questions, right? I'd like to, um, if you're early into your career in the first, you know, this is your first startup, you have liquidity matters a lot.

It's highly underrated. But, uh, so my advice would be have those early conversations with people. And if you don't feel comfortable having those connections. That's about as big of a red flag as there can be. It's like, you just don't have the level of trust or get, yeah. Just get to know them. Um, it'll be better for everyone. 

Brett Berson: when you reflect back on those fundamental question conversations you and Brian had, did the answers that each of you shared, uh, ended up being the way that you behave? In the past 15 years when those questions came up or there is a little bit of a trick of like my current self and future self or are related people, but maybe slightly different

Dharmesh Shah: a mix of both, but a couple of them were like fundamentally like deterministic of the trajectory of the company. Um, so for instance, one of the questions we had early on is around this, you know, what does success look like? What are we trying to accomplish here? And, uh, you know, both Brian and I had some kind of success in our, in our professional careers.

You know, I had started before I had sold it. Um, so I had hit it like a single or a double. And the only reason I was doing HubSpot is because, um, you know, I hadn't quite ever hit a home run. It hadn't had anything big and impactful, uh, done that. I felt like I had needed one more step up to bat. And Brian was in a similar position.

So one of the early decisions we made was, uh, what we're trying to do here is build something sustainable, build something big and impactful that, um, you know, someday our kids and grandkids, uh, we didn't have kids at the time. Uh, we'll be proud of something. We can kind of play back to them. That was the aspiration.

And we also talk about doing it in Boston. but once we made that call early on, this was literally like week one of the company. Um, out of that one decision. So many decisions became clear, um, which is every time we hit a fork in the road that says, okay, we could go down this particular.

Um, and then the question we ask ourselves is which path, um, maximizes the probability that we have a chance at this kind of big, impactful, significant company. Um, and so we never took the conservative path. So when we were out raising capital, we didn't worry about dilution. We worried about it, like, okay, does this improve our odds of being the big impactful company that we want this to be?

Uh, and that was super useful. So we didn't get distracted by acquisition offers. We didn't get distracted by trying to negotiate the best possible terms and, and worry about what dilution was going to be. We needed. We wanted fair deals, um, you know, with all of our, uh, private venture rounds, but it made a lot of things easier.

Um, as we were trying to maintain control, we were, it's like, okay, well, let's build a great board. We knew, uh, we were not reluctant to raise funds. We were not reluctant to go public because that was all part of. Does this help us make, turn us into the big, uh, significant company we want to be? Does this help us kind of achieve our ambitions and our aspirations?

So now. Question that we asked each other in that, uh, that first week I had that big of an impact, but many of them did. Right. And that was one of them. 

Brett Berson: do you think there's a difference between your personal values and Brian's personal values and the company. Core values or is the company's core values, just an expression of, of the way that the two of you behave.

Dharmesh Shah: that's a good question. So Brian and I are probably 95 plus percent overlapped on, um, just core values. We, um, many of, uh, joked that Brian, I kind of think with one mind almost, uh, in some like I can, and it works in both ways. He can predict how I would respond to any particular question in any particular situation.

And we've had our kind of disagreements on a tactical level, but our values are, are very, very similar. Um, and I would say, you know, the culture is, so if Brian and I are 95% overlap, the culture is probably maybe 85 or 90%, um, overlapped with the two of us. Um, Lot of the core of the culture kind of manifest as a result of just how bright I think of things.

Uh, but you know, the cultures do drift a little bit as you add more people. Uh, it becomes harder, hard to try to keep it, uh, centered on one particular thing. And I don't know that that one particular thing is necessarily the right thing. Um, and so I think, uh, you know, we can talk about this a little bit, but you know, I think of culture as a product and, you know, product is not static.

It's not like, oh, we have this culture, you know, the early years of HubSpot, didn't that work great. And now our job is to preserve that culture. I don't think that's the right approach. You don't want to preserve your culture. You want to iterate on your culture, keep the fundamentals, the thing that makes the product, the product, or the culture, the culture, the same.

Uh, but like it's not a static thing. You have to iterate because the needs of the market, uh, change the needs of the customers. So we think of culture as a product. And by the way, when I say we think of culture as a product, we literally think of culture as the product and all the things that would kind of manifest from, from that one single, uh, fundamental belief. Sometimes to play of ludicrousness. Uh, but we, we treat it that way. So we think of the employees as customers I've said product, and that's why we do the surveys as if we were doing a product survey. That's why we follow up with, oh, here's what our kind of roadmap looks like for the culture. Here are the things we're fixing or not fixing, um, that all kind of traces back to this notion that a culture is a product that you build for your people so they can get, um, make good decisions and help grow the company. 

 one of the things that I was reflecting on in our conversation thus far, is that it's quite interesting that you're tend to be more introverted, um, prefer to not manage people. And yet you, you seem to have this intense curiosity and excitement around culture and people. And I'm curious, where does.

Brett Berson: Comes from given, it feels a little bit antithetical to some of the, some of the ways you described yourself.

Dharmesh Shah: Yeah, that's interesting. So I'll tell you the it's, it's a good story, um, of how I, so I'm informally responsible for culture at HubSpot, um, and, or LLC And unofficially it's actually somewhat formal. Um, but here's how this turned out. So my co-founder, uh, was part of a CEO group that, you know, they got together quarterly and talked about CEO issues.

Um, and I always pictured when they had these meetings that they'd be like out in the field somewhere sitting in a circle and singing CEO songs. Um, and the topic at one of these meetings that Brian went to, um, was culture. And this is, you know, early, early years, maybe you're a four or five of HubSpot. Um, and.

At this meeting, you know, culture was a topic that they would go around and everyone would talk about lessons learned on culture, what they're doing about culture, so on and so forth. And he had some highfalutin CEOs as part of his group. And so then we, and I wasn't part of this group, I hadn't been part of this meeting, but we had one of our, um, regular founders dinners and, and Brian kind of caught me up.

It's like, yeah, I had my CEO group meeting a couple of days ago and the topic was culture. And, you know, I told them, it's like, well, you know, we're still in the very early stages. We're like building product and selling product and trying to figure out how to scale. And culture is a thing that, you know, we will definitely address, but that's going to have to come later because it's just not, you know, it's not a burning issue right now.

And the rest of his CEO group kind of came down a little bit hard on it as LeBron, you don't get it like culture is the number one thing. If you don't get that right. Or if you mess that up, nothing else will matter. That's the thing that's going to kind of determine your long-term destiny. And so he's sharing this feedback with me.

It's like, so culture is important. I'm like, okay, Brian, that's awfully good insight. Those are smart folks. Great. And then his following sentence was, and so why don't you go do that? And I'm paraphrasing. And I'm like, do what, like the culture thing, like go figure that out. Like of all the people in the company to work on culture, you're going to pick the one, that's the least social and likes being around people, the least.

Um, and I'm one of those it's like, okay, that doesn't seem like it makes a whole lot of sense. Um, but you know, I want to be a good co-founder. He had lots of things to do as, as, as we know. Um, and so I said fine. Um, I'm not exactly sure what that means, but fine. And so then I treated as a, being an engineer, uh, treated as an engineering exercise.

So like step one was I'm like, okay, well we have a culture now. Like we do things. We make decisions as a company is running. Um, it looks like it's going pretty well. And so like treat as a data exercise or analysis exercise, which is okay, we already have a culture let's stipulate that it's working pretty well.

Let's figure out what that culture is. And my job is really to kind of write that down and kind of manifest it somehow. And so they went and did surveys, um, of all the employees to figure out. And I did it the same way I would do anything. I did a net promoter style, uh, survey. It's not a scale of zero to 10.

How likely are you to recommend HubSpot as a place to work? And the quality of the question is like, why did you give that score? And what came out of that? Well, it's have to share this kind of mini story within a story. So the first time I reached out to the team at HubSpot and said, Hey, I'm going to be like trying to figure out our culture because Brian, I had this meeting and we think it's important.

I forget how I said it, but it was kind of like. And in the 15 years I've had it HubSpot the most negative visceral reaction from the team I've ever gotten of anything I've ever said was that declaration that I was going to go work on culture. And it just kind of took me back a little bit. It was like, I was like, okay, well, where is this coming from?

Like, what is this? And, um, and, and so I talked to a few folks and by the way, they're like most of these people I had hired, and then one of them was just kind of forthright with me. He's like, and I, I asked him like, why are people? And you like feeling so negative about the fact that all I said was like, we're going to look into our culture and figure out what it is.

And he's like, well, you know, that's how, that's how these things start. We're going to look at culture and then what's going to happen is we're going to have these posters on the wall with our mission statement and our values. And we're going to become one of quote, unquote, those companies and, uh, another long-time person.

It's like, okay, well, I don't know if HubSpot's the company that I thought I joined. Dude, how can that be? We haven't even done anything yet. Um, but I was still, I was super sensitive to this, um, kind of visceral reaction. People have negatively around this idea of going to crafting culture defining culture, because they thought it was a, should be a organic thing.

It's not something you actually work on and disorder is and they may have been right. I don't know. Um, so I kind of treated the, treat it a little bit gently. And so I, so I made it this kind of data collection exercise. It was simple set of questions, just trying to figure out the culture and, um, and that first kind of deck, um, that kind of described it was called.

I called it the culture code, um, and that names still exist today. And most people, when I talk about the HubSpot culture code, they think code is like a code of conduct or a code of values. And it's not that at all, it's a code I meant like literally code. And so the kind of V1 of my project was if I could write a funny.

To mathematically approximate the probability of the success of any given HubSpot person. What would the coefficients of that? What would the, uh, what's the parameters of that function? What do you pass in, uh, without figuring out that the relative weights and such. And so that's what I thought of as my exercise.

I just want to identify the things that are correlated with success at HubSpot. And so that first version identified that, and transparency is one of the things that came out of it. Humility is another one that came out of that early exercise that people tended to be humble and valued the ability and others that HubSpot who knew I, that was not something I had really thought of.

And so subsequent to that exercise, uh, that deck stood for a couple of years. Um, and then people can. The team then came back to me. It's like, okay, well that is actually very useful. Um, that first deck, because it helps us hire helps us. It's, it's, it's a useful thing, but it could be more useful because all it really tells us is like the who, like who kind of fits it, HubSpot who's likely to do well.

It doesn't tell us about how we should operate or how we should make decisions. And so they added asked for kind of new features to this, uh, culture code, uh, this operating system that we were trying to write. So I went back and, uh, so the original deck was 16 slides, uh, went back and expanded it, uh, to 64 and then ultimately 128 slides, which is what it stands at today and will not change because now every time I add a new slide, I take a, another slide out.

But, um, so anyway, that's how I kind of fell into culture. So now it's kind of worked in a way, oddly paradoxically, well, because part of what makes me use. Is that I look at culture in a somewhat of a scientist's way, like a size scientist would like study a grasshopper kind of thing. Right. Cause I have no horse in the race.

I have no team I'm trying to protect. I have no direct reports. I can have a somewhat more objective view on culture and how things are working. Um, just giving them the unique nature of my role. And obviously I have lots of skin in the game and lots of incentive and motivation to kind of try to make the right calls.

Um, so it's, it's worked out, but that's um, had that early kind of CEO meeting between Brian and his CEO friends not happened. I don't think I would be as deeply into culture as I am now. Um, but it worked out.

Brett Berson: And this idea of culture is product. What are cultural products that you've shared? That failed. And maybe what are products that you shipped that had an impact beyond your expectations?

 if you think of the HubSpot culture as a product, what are the key features that, that people ask for that they want? Right. And autonomy is way up there. Transparency is way up there. Uh, but, uh, one that has kind of shown up over time, especially recently, but even pre pandemic is, is around flexibility as a core, like literally a core feature of the product.

Dharmesh Shah: Um, and we did okay on this front. Uh, but, but it kept with. That drum kept beating. It kept showing up in the, in the NPS surveys that we did. So we've been kind of expanding that particular kind of group of features around, you know, how do we make HubSpot more flexible? Um, along several dimensions, one is obviously around just remote work and the decision around being in the office or not in the office.

 you know, one kind of attribute of the early HubSpot culture was that we didn't have titles because we didn't. Cause we think we thought titles reflected hierarchy and we want it to be this kind of meritocracy based thing. And we thought kind of titles were stupid.

Dharmesh Shah: That was the original, um, original thing. And ultimately we had to kind of retract that, um, just because the customers revolted, um, and we introduced the notion of titles and we had three options on the table. We said, okay, well we can continue to know titles. So that was kind of option a, that was a default option.

We could do, uh, titles that were kind of classic titles manager, director, VP, senior VP, um, or we could do titles, but everybody gets to make their own title up. So you can be grand Poobah design or something like that. Um, and ultimately we chose option two, which was classic titles, um, because the customer feedback, employee feedback.

Well, the reason we want titles is because there's life outside of HubSpot. Um, rumor has it. And that if, as we go on with our careers, that's the thing that people are going to look at as proxies for progression, uh, in our career. And that's what family is going to ask at when we get together for, for the holidays, like, oh, like, you know, how's work going, whatever.

It's like, oh, I got promoted from a director to a, uh, to a VP or something like that. That means something to the outside world. Even though it may not mean that much at HubSpot. So that was one of those that even though Brian, I felt strongly about it, this kind of notion of meritocracy and that titles were kind of silly.

Um, we, we, I think rightfully got convinced that that was a, uh, that was suboptimal. And so we made that, made that change.

 I'm going to share another early story with you, um, from, from the early HubSpot days, so in our first office, um, and we were in a coworking space and we had a room devoted to ourselves, um, and it had. And there were only two of us, my co-founder and I, because, you know, we had big aspirations to grow.

Dharmesh Shah: So we had twice the capacity of the existing, uh, employee account. and so, and I just want you to visualize this room. So there's a room, it has a window, and there are two desks that are by the window and two desks that are not by the window. That's the only, uh, distinction and criteria.

There are kind of four corners of the room, not a very big room. Okay. So we hire our first employee, uh, senior engineer. He was, uh, out of Yale. And so we had to make a decision around just seating right now. Like the first thing that came to us was like, well, you know, we don't really care all that much about like the desk.

We just happened to be sitting by the desk by the window, because why not? Um, just away from the door away from some of, uh, kind of hallway noise, but it's like, okay, well, it doesn't seem fair. Um, for. As to I've already occupied these two seats and maybe Patrick, the new guy wants to have one of these seats and like, we don't care enough anyway, that's not what this is about.

And so we did this weird thing we said, okay, we're going to, uh, um, do this like a mini lottery, right? So we put, uh, our names into a, uh, into a hat. We didn't have a hat, but, uh, and then we picked out whoever's name got chosen first. And this was a key part of it is they got to choose of all the four desks, which one they wanted.

Right. It wasn't like, oh, whoever gets, you know, they get the window desk or what we believe to be the ideal optimal desk. It's like, no, you get the choice and you can pick whichever one is optimal for you. Uh, and so we, we did that. And so, uh, he actually picked up one of the model window desks. And so we went back and then the next hire, we did the lottery again.

The next time we did the lottery again, we, and so we kept doing that as we, even as we moved into larger and larger office space. And, um, there was a couple of things. Kind of manifested as a result of this really simple hack, right? Like any company can do it. Uh, one is, we hadn't really thought about it at the time.

Um, but it's like, oh, people got to know different people because you were never locked into one spot. So, um, and if you've got a super crappy desk, it was only going to be crappy until the next tire, um, which happened relatively quickly. And then we did it on a quarterly basis. Uh, but the big, big, big value out of this is it took so many cycles and calories that people would have possibly spent on thinking about their desk or their office space, which happens in bigger companies.

Uh, they just didn't have to spend, because that was a literally a non-issue no one had to think about it. Cause they were all the desks were identical. Um, and it was done a lottery basis anyway, so it didn't really matter. And if one, you know, happened to, you know, not get a great desk one time where you could, because you'll have another lottery in a, in a three months anyway, and that was awesome.

If I could go back and find other hacks and this titles were part of that. And so these kind of investment in culture, our hope is that,

we can make the operating systems such that, um, like the people in the company are spending minimal cycles are no ops, you know, things that just don't instructions that don't accomplish anything.

 just to kind of close out that story. So, so we did it at, obviously at two people, did it, four people did it, eight did it six, we kept doing it. And then we had like a real board on, in place because we had done some venture rounds and the board and they would, and we would talk about this kind of jokingly and they're like, okay, that's cute.

Dharmesh Shah: And that's all fine. But you realize that's just not going to scale, right? Like that, that can't work every three months changing everybody's seat around. Um, so that we did it a hundred, did it at. And we kept doing it and we do it to this day. Uh, and we made some tweaks to the algorithm, right? So it's not a complete random distribution anymore.

What we try to do now is to say, Hey, we want to try to keep the quiet people with the quiet people, because an engineer's sitting right next to a salesperson that's on the phone all day is suboptimal. so we still get the kind of positive effects of the randomness and the fact that people don't have to spend calories thinking about which desk they got.

We also get the positives of people get to be around other people. Um, and none of the negatives around the politics and the, the, the, the kind of meta gain calories. So it's, it's, it's worked out, we still do it. Hundreds of people will change desks every, uh, every three months

Brett Berson: well, that's a great place to end. Thank you so much. Uh, for this conversation, it was, it was a blast. 

Dharmesh Shah: My pleasure. Thanks for having.