Andrew: There's a difference between companies that are like, okay, I'll get back to you next week versus I'll get back to you tomorrow versus I'll get back to you later today. The team that has like just the faster cycle time, they may not get it right the first time, but they're probably getting more shots on goal and ultimately be more successful.
Logan: Welcome to the Logan Bartlett show on this episode. What you're going to hear is a conversation I had with CEO and co founder of Klaviyo, Andrew Bialecki. We talk about a number of different things over the course of this conversation, including building an entrepreneurial mindset among your employees.
We've taken the same kind of like C, C, A, B, C model and just set that up internally. We'll expose to them like they're P& L. This is how many people are working on this project. This is how much you cost. This is how much revenue you're bringing in. We found that people are very motivated by that. Like ownership of an urgency around what they're trying to accomplish.
Why there's magic in obsessing over details. Why culture is a product that you actually ship. As well as his philosophy on fundraising, which included nearly bootstrapping the business for a while before raising significant outside capital. Klaviyo has grown into a six billion dollar plus public business.
Andrew has a number of different lessons that are helpful to any entrepreneur. that you'll hear now.
Andrew, thanks for doing this.
Logan: So
Logan: I want to start off talking about the benefits of constraint. And I think either by your own, uh, Uh, decisions or maybe initially, uh, the venture capital ecosystems decisions, uh, there were constraints that you either chose or were forced on you. And I think, um, I think that that those constraints, it sounds like benefited you guys in a number of different ways, maybe related to focus.
But how do you think about constraints and like? The advantages around them and how it's benefited, you guys in your journey,
Andrew: Yeah. So, uh, well, first, I mean, what constraints were kind of like born through constraint. Um, so, uh, you know, Cleo was a bootstrap business. Um, we, uh, we founded the business in 2012. Um, I remember [00:02:00] quit my job, you know, cornered my, uh, then was going to be co founder ed. He was a business school and said, Hey, like, let's start working on things.
We start working on it. Um, well, he was graduating from business school. Um, and then it was,
Logan: wasn't doing much anyway at that
Andrew: well, I had a hot, so one of my, uh, he can advocate for an MBA, but I'm a very, I'm a big believer that MBAs are a waste of money. Like, just go work at a startup, like you'll get paid, uh, so you hopefully can save up a little bit.
Um, don't have to take on the loans and the debt and you'll learn just as much. Right. Um, and I'm a huge reader, so I kind of feel like if you want to read the case studies of the business books, like you can, you can do that in your spare time. Um, but anyways, yeah, so we started in 2012 and we didn't raise any venture capital until, um, you know, we were already profitable.
Um, and I think it was 2015, 2016. Yeah. So, uh, we're kind of born out of, uh, constraint. And um, you know, before I hit on that, like a lot of people ask, like, well, well, gosh, why didn't you do that? And there are really like two reasons. The uh, the kind of, you know, the um, uh, the one, one was, you know, and I had both come from families with small business backgrounds and I always felt it was kind of weird that technology companies were the one companies where it's like, yeah, it's okay to just be a cash incinerator.
Uh, but every other business is like, no, you have to turn a profit, you know, maybe not day one, but you can't wait that long. So and especially, you know, the marginal cost of building software, I mean, it's like, you know, we had, you know, there was cloud compute and AWS was a thing at that point. Like it just was way easier than it was in the 2000s.
So I was one is like, we felt like it's possible. Um, and the second one was, uh, so I mentioned, you know, Ed was at business school. He was at, um, uh, he was in, uh, at MIT. And, uh, you know, all the, a bunch of the, you know, private equity venture capital groups would do these kind of summer at X. And so they had these programs where they give you 20, 000 for the summer and you had to put together a little pitch video and, you know, they just give you like kind of like, you know, they give you a, uh, no, no equity funding and, you know, a place to work.
[00:04:00] And we thought that was awesome. It was 20, 000 bucks. I got some fun for the next year with that. Um, we applied to like three or four of those and got into none of them. Uh, despite the fact we're like, we have a product, we have customers and, you know, so I remember telling you that I was. If we wouldn't try to pitch, you know, the actual, like, you know, VC fund, we're like, gosh, Ed, we're like batting zero.
You know what I mean? I mean, maybe we're just not there yet. So anyways, you kind of took it as like, all right, we're going to figure out how to make this work with just the two of
Logan: What, in retrospect, what was, did, was the articulation not
Andrew: I think so. Yeah. I mean, we had this. So we had this point of view that we were just going to obsess about products. You know, my background was, um, in software and a little bit in product design and Ed was in product design. Um, and, uh, we said, like, look, we're gonna, that's about building product. We thought we were really good at that.
Uh, and we were just gonna, we both believed in, you know, lean startup was a thing at the time. And so it was getting close to customers and we're gonna find real revenue. And so, yeah, we just put like no time into, you know, the pitch pitch decks. And yeah, I thought it was funny because, you know, folks be like, well, I don't really get it.
And I'm like, well, here, I've got like, got like 20 customers, you know, paying us, I don't know, a thousand bucks. Like, I mean, there's, there's a, there's something here. Um, but what we just didn't, we didn't invest in that part of it. So it wasn't actually until we did some more formal, you know, fundraising, I was like, okay, we probably should like clean this up, um, the actual
Logan: So some of the constraints were forced on you, or at least you were guided towards
Andrew: We like to, you know, the, the, the rosy picture was, you know, well, we're going to do this because this is the right way to build a business. And then there's this other side of like, okay, well, maybe there are, maybe there are no other capital sources. So we just got to figure it
Logan: And you were technical by background and most of the at least the biggest costs of starting tech companies are usually hiring developers to go do it. So you could you could do it yourself. Uh, and execute on
Andrew: Yeah. Another, another thing I advise, uh, you know, founders are, I tell folks, it's like, it's like, look, if you are a non technical founder, then you should go find a technical co founder. Um, if you are, you know, [00:06:00] two or three technical co founders raise as little capital as you need, Um, you know, one of the good things, this is not really a constraint thing, but we've had this rule inside of Klaviyo that we don't celebrate fundraisers.
Um, so, you know, other than our, other than our IPO, which, you know, we're like, yeah, we'll have a party for that. But yeah. Otherwise, we're like, well, it becomes kind of a Vandy metric. It's like, well, nobody really cares. And it's, you know, it makes you feel good because it's a little bit of validation. Um, but we had this point of view that it's like, you know, we're just not, we're not going to, we're not going to focus on that.
Logan: Yeah. No one celebrates, uh, taking out mortgages on your house. I don't think,
Andrew: Yeah, exactly. Same mentality.
Logan: it's sort of like you're, you're giving up. I mean, there's, there's elements to be celebrated that there was outside independent validation and you know, your parents say congrats or whatever your, your, your spouses think that you can stay employed for longer, but it's not inherently in and of itself a point of validation on the actual terminal value that you're going to create.
Andrew: that's right. That's right. That's right. So anyways, back to constraint. Um, so when we started, you know, one of the great viral loops, um, in any product company, uh, technology software company. We always felt like there was, there was, you know, there's kind of building and there was selling and we used selling as like really like just working with customers, right?
Doing the, you know, okay, well, what do you think of what we built? How could be better getting them to adopt it? Like figuring out that kind of, there's a real great flywheel back and forth between those two. So I'll give you like there were a bunch of great patterns that we developed and then carried forward.
Like for instance, one was, Um, we had this shared, you know, um, email inbox that we used to do support. So after we got a couple dozen customers, you get questions, you know, uh, you know, every day, every week. Um, and we had this rule that every morning we'd wake up and we weren't allowed to work on anything else until we'd like cleared that inbox.
So you know, that was especially interesting as we got to, okay, you know, you've got a dozen, you've got a couple dozen. You know, when something broke, [00:08:00] you heard about it. If something was hard to do, you heard about it. And you know, for me, a lot of my days were, all right, I'm going to solve for that stuff in the morning and then I'm going to go code and build the next, you know, the next stuff in the afternoon, right?
In the evening. And if a lot of my day was sucked up by responding to these questions, I was like, okay, well, the first thing that you'll fix is whatever those problems were. Uh, that create a lot of great discipline about building good products that it's interesting. Like as we've grown, we've had to work hard to kind of keep that loop tight because it's natural that, you know, if you, if you split up into, okay, well I have a support sport team or a customer success team and I have a product engineering team, it's easy for there to become this like kind of like, you know, buffer zone, right?
And then those things just don't get resolved as quickly and the product actually suffers for it. So. Yeah. We think a lot of the reason we were able to go so fast was because that loop was super tight with us. Cause it just, it had to be
Logan: There's the cultural principles of that, which. You've tried to maintain and we'll go into, uh, I definitely want to ask questions about like how you maintain that over time and as you scale and when you get to hundreds of people and hundreds of millions of revenue and all that stuff, which I think it's interesting.
But then there's the, there's, there's like very counterintuitive element about, uh, what What focus, uh, breeds and that. And so like finding your end market, uh, as potentially an accelerant to, uh, what you you've done or limiting head count. And so there was less, um, uh, silos of communication, which is partially tied to what you just said, but maybe a little distinct as well.
Can you talk a little bit about that?
Andrew: Yeah.
Andrew: So, you know, I mentioned that, you know, my background is like building products and the idea for Klaviyo, um, when we started and still is today is I think the future of the way that businesses and consumers interact with each other, you know, I grew up in the, you know, the nineties and I remember when it's like, if you wanted to go deposit a check, like you went to a bank teller and there was a person you had to give it to.
Uh, [00:10:00] you couldn't deposit an ATM, you could take a picture on your phone. Uh, if you wanted to change something about, you know, if you wanted to add, uh, HBO to your cable bill, you'd have to call somebody, that wasn't something you could just do yourself. So everything was, the idea was, if you're a consumer of some service, you have to call somebody, you have to interact with a person.
Okay, but our belief is that is, you know, the old school, it's going to be anachronistic, everything is going to be mediated by software. And so then the question is, is like, okay, well, how do you build software that can be You know, automated version, the scalable version of a business or of a person. So I like to think like whether you're a solo entrepreneur or your business, there's kind of this, this, you have this brand, this way you want to interact with the world.
And we look at Klaviyo as that kind of surrogate that makes you infinitely scalable, right? We talk about being yourself in an internet scale. So we're that scalable brain for business at scalable, um, voice. So that was the concept behind Klaviyo. So when we built it out, it was, it was very abstract. Um, we actually started, you know, we wanted to sell to other startups.
So we did what every good, uh, uh, startup did back in the 2010s. And we post on Hacker News and try to get a bunch of startups to use us. Once we had the product going, the next question was like, well, where are the, where's the demand going to come from? Where are the customers going to come from? And we looked at it's like, well, there's tons of businesses that could use us.
We just have to figure out where the right pockets are. So, uh, it's interesting. The startup thing worked okay, but we found that there were a lot of, a lot of software businesses that didn't really know, you know, how their customers were supposed to use their software. So there's a lot of consulting we had to do and it was very slow.
Um,
Logan: And would that have been more B2B
Andrew: yeah, exactly. Yeah. Yeah. I mean, we're kind of selling to businesses like us. So you think of like the normal onboarding flow, somebody signs up, you say, all right, so what are the actions that make this. It's a person successful with your, you know, with your software. And we'd, you'd go talk to a bunch of, um, founders or teams and like, well, that's a really good question.
We have to figure that out. Can you guys answer that for us? And you're like, oh, okay, gosh, like you guys haven't even solved this problem yet. Um, then a friend of mine had a, uh, had an e commerce business, uh, and he started around the same time as us. [00:12:00] And, uh, of all things, he was, he was, he was creating, uh, t shirt quilts on the internet.
So you'd, you'd send like, you'd literally collect like old t shirts, right, say from high school or college or, you know, activities you're a part of, and they would collect them for you and turn them into like a blanket, right, into a quilt. And so he's like, yeah, well, I'm on this, uh, this platform, Shopify, like, and, uh, I was like, okay, well, like, do you think this is like relevant to, I mean, you've got all these, it's pretty customized products, um, and then a lot of repeat purchases, right?
Like, you know, you end up with like a parent that would buy for their kids as they graduated college and all this kind of stuff. And he's like, yeah, totally. Uh, and so we started there, there was another business, you know, so that was one business. We found a couple of other ones in around the Boston area where we was, where we, um, found a Clavio and, uh, they all had the same thing.
And so our goal the whole time though, was how do we get like scalable demand? So after we'd kind of hustled for those first 10, maybe in retail and e commerce, um, we had this idea that was, gosh, we need to find a way to get people to come to us and we don't have a lot of, you know, dollars to spend on advertising.
And so we had this idea that, well, gosh, there's, you know, sass is kind of becoming the thing. People were shifting away from open source and hosting self hosting to software as a service where they have to set up the software. We said, well, what if we just go build partnerships with those platforms? Like, what do we say?
Hey, look, we kind of compliment your software. And, you know, we kind of, we complete this picture. Maybe that could give us a way to get, you know, at the time for us would have been like hundreds of businesses to come to us, um, you know, in a week or in a month. And then at that point, I was like, okay, we have enough customers, we can keep go back to work on the product and kind of expanding it.
And that strategy worked great. Um, because what we initially built was, you know, almost like we built, you know, uh, this database and marketing software, but you had to connect all your data through these APIs. And we said, what if we just go partner those platforms, we build all these connectors, kind of build once run, you know, hundreds of thousands of times.
And [00:14:00] uh, that strategy worked great for driving demand. And so that's one of the ones where, you know, I think a lot of, let's say software companies, a lot of them still, you know, it's either, it's either just find the call list and go start calling down, you know, uh, start calling folks or it's, uh, do advertising.
And I think we thought a little more creatively about it, uh, really about this idea of partnerships Transcribed Um, and how you could use those to do, uh, you know, awareness and demand gen. And it turned out to be really scalable and also really efficient. Um, that's one of, you know, the first kind of little S curves that, you know, we took off
Logan: And so, so that took you into e commerce and retail as kind of an in market. And at that point in time, so now you're building distribution, but you're, you're opting into these partnerships that, uh, service a narrower sub par. segment of the overall economy, your tools broadly applicable. But now, and so then you're facing the decision, I assume, of like, do we tailor our website copy?
Do we tailor the product considerations a bit more to that?
Logan: And so maybe maybe talk through that decision tree of like, okay, well, we have this tool that can be broadly applicable. And we're actually narrowing our in market in some ways,
Andrew: Yeah. I remember it was, I would, uh, uh, fights probably too big of a word, big, big debates. Uh, my co founder and I had where I remember our homepage, I had, we'd very intentionally designed it to be very generic, right? Klaviyo, we're the brain for your business, you know, store everything you know about your customers and then like, you know, treat them the right way, message them at the right time.
And it importantly, it didn't have any of these things of like, oh, there's nothing for e commerce or retail on there, but we were starting to get this growing swell of e commerce businesses that were coming to us. And we had these great quotes from customers like. Uh, you know, I remember there was a, a backpack company that said, Oh, you know, other softwares for open clicks, Klaviyo's for sales.
And it was like, Oh, great. It's like the best tag. We should just put that on the website.
Logan: which is great about e commerce [00:16:00] is like, there's actual tartar or why that you can
Andrew: Yeah. Yeah. It's very direct response,
Logan: not B2B sales where it's like, yeah, we got, you know, 5 percent better leads coming
Andrew: I agree. Yes. Yeah. Yeah. Yeah. Totally. Right. Right. There's the conversion cycles. Very quick. Totally. Um, so I remember Ed and I give him all the credit for this, you know, kind of games you one day we said, he's like, uh, you know, he's like, uh, I guess, uh, weekly roundup staff meetings when there were like three of us.
Uh, but it was like, it was like the two of us, we'd sit and we'd have a, uh, a beer and pizza at the end of the day. Uh, at the end of the week on a Friday and he's like, look, Andrew, I really think we should, you know, I think we should change the website and really focus on this vertical. We're getting a lot of traction there.
We'll just go faster. I know how much you care about going broad, but like we can go do this and then, you know, make it more horizontal. And I remember going home and thinking about it for the weekend and being like, you know, kind of grumble. I really don't want to do this, but he was totally right. And so we did it.
We did the next week and it was 100 percent the right decision. Um, and, uh, yeah, I think, you know, we had not done a ton of market sizing at that point. I mean, we knew it was a big market, um, cause we'd studied some of our competitors and like where they're, you know, who their customers were. Um, so it was plenty big enough to say, get us to, you know, 100 million in run rate.
Um, but I think, you know, we've been, uh, you know, as we've studied it more now, it's, I mean, it's obviously a growing segment. Um, but retail is just such a huge part of the economy and, uh, and we also realized how underserved it was. Um, so going vertical really helps. The other thing that really, I mean, I, part of what Ed pitched me on and part of the reason I thought this was a great idea, uh, was if you looked at our competitors at the time, whether they were kind of the traditional newsletter type companies or, you know, some of the bigger B2B SaaS companies had their kind of marketing clouds, marketing suites, they were
Logan: on one side and the Salesforce marketing
Andrew: right.
So they're all like, and they're all very horizontal. And we did think there was some kind of arbitrage of, you know, we were all about, okay, we're going to do more [00:18:00] personalization, more targeting, we're gonna be more focused on conversions and return. And you kind of do want to go like vertical by vertical use case by use case for what does that mean for you.
And we felt like there was some arbitrage of okay, they're going very wide. If we focus in one area, we can kind of, you know, go win that market. And then we can take that playbook and do it for others. And in fact, that's, you know, we've got a whole team internally. That's what they're doing is they're basically standing up effectively, you know, industry by industry.
So I tend to think that like, you know, when you, when you, um, when you, when you pick which markets you're going after, uh, I think generally speaking, you should try to build product that is horizontal. I mean, there's obviously a lot of judgment there and how wide you go. Um, but something that's a little more abstract, but you have a lot of degrees of freedom on the, you know, marketing and go to market side.
And I think you have to choose whether you're going to go horizontal, which can work if it's kind of a newer category. It's a little bit of like, Hey, everybody needs this and there's potential for lots of lots of demand. Um, or you can go kind of vertical by vertical in either. Either way, I think once you get to scale, you end up with this, you have some sort of, you know, uh, horizontal product marketing and then an overlay on top of that.
That's kind of like, okay, use case by use case, industry by industry. So you can get to it in two different ways. And I think in our case, I think one of the really smart things was we're like, well, because everybody else has gone broad. If we focus more, you know, vertical by vertical, we'll stand out relative to them.
That'll give us a little bit of a
Logan: More resonance with the end customer. And I guess I don't want to speak for you, but I would guess that had venture capital been raised or a meaningful part of, of, you know, the underlying business consideration, both the distribution of how you were thinking about going through partners and then also the narrowing of the TAM.
I, I assume you might not make those two decisions on the same tree. [00:20:00] It's like, okay, we're going to, you're, you're going to control less of your own destiny from a accelerating growth standpoint, and we're going to narrow the TAM that you're going to target. And so both those decisions, I maybe would have come at odds with venture capital.
Andrew: I'll say like, yeah, I mean, two things. One, uh, you know, it's hard to start a company. Um, it's hard to start a company and cause there's so many decisions you have to make, right. And it's, uh, and you know, there's a lot, you know, everybody's staring at the mirror is like, you don't always have 100 percent confidence that you're gonna be right about those.
Right. So one pattern I've seen from some, some auctioneers founders is they sort of want some help making those decisions. And I think they can over fits to their investors, right? Because investors, I mean, really great ones are great advisors. We've had some excellent ones. Um, we had like, for instance, we had very good say, investor or founder fits with our, um, uh, seed investors.
Um, yeah, I think that's right because I think there's generally speaking, uh, a lot of venture capital investors would say like, well, look, bigger, tan better. So don't narrow things if you don't have to. And that might be right. It's just, there's some nuance to it. And I think people want helping decisions.
So sometimes you kind of have to have your own point of view and not wait for other people to tell you what the right answer is. Um, what was good for us on that front is actually our seed investors had a bit of a contrarian view on that, which was. Hey, I actually think there's room to start going, you know, kind of vertical by vertical and that you can build a big business that way.
Um, this concept of vertical sass was like, you know, I'm saying newer 10 years ago. Um, and I think that was very much to our benefit because like we're actually very much like on the same page. Um, the other thing I'll say is I, you know, I, in the last 10, 15 years has been a lot of software companies that I feel like have, you know, they've, they've kind of their ambition has gotten smaller.
Uh, there's been a lot of like. Unnecessary nichification of, well, I'm going to go off this one little thing and that's gonna be my [00:22:00] pocket. It's a fine place to start, but I do think you need to have the broader ambition of, okay, well, how is this going to scale out? Like I'll give you one example that was relevant to us and our product.
You know, when we built Klaviyo, um, this, this notion of like conversions, like what happens after you run a marketing campaign? Well, what was interesting was we saw a lot of other companies that even tried to do something similar to us, but they really overfit to this idea that like a conversion meant like a retail transaction.
And that was the only thing that would ever matter. And we were actually able to take a lot of market share from those folks because they had so pinned themselves to this idea that the only thing that matters is this conversion event, right? And what we found from retailers is they're like, well, sometimes it's like getting somebody to sign up for a loyalty program that matters.
Or sometimes, you know, we're seeing increasingly a lot of retailers where retail is one part of the business, but they also offer services. Um, and so keeping that more abstract. Um, in generic made a, made a big difference to us. So I think there's also like a little bit of, you have to know the game plan, right?
Is the game plan to start vertical and then go broad? Or is the game plan just to be vertical? And in that case, I actually think like for us, we, you know, we, we knew that it's like, okay, we're gonna start here, but that's, we'll never be the end goal, right? So the ambition is still big. So that's the other thing I kind of look for in companies is do they have this greater purpose?
Is this like a stepping stone? It's like part a. Absolutely. Of this like greater master plan or is that like the whole
Logan: Yeah, and I guess at the time you were getting going too, it was a lot of I mean, there's the Jim Barksdale quote of, you know, bundling and unbundling. And, but, but at that point in time, there was a lot of best of breed niche functionality. And now it's probably come a little bit more in vogue. I think Parker Conrad started the compound startup as like the, the adage around it, but being broader in your ambitions rather than we'll Let us take the ATS part of the market and someone else will take the, you know, the LMS side of it or [00:24:00] whatever, all those different things that I think you guys have done a great job.
HubSpot, uh, has, has built out like a lot of these different modulars within a broader suite. I think one of the, I mean, this is a total separate thread on this, but it is interesting when I get asked the question, uh, I'm sure, sure you do as well, but like the, the, uh, someone, uh, 22, 23, 24, five has the question of like career advice, right?
And, um, And it's counterintuitive to tell them to be like, no, no, no limit your options. I'll get a lot of ex bankers or consultants or whatever. And they're like, How do I get a job in investing? And I'm like, the best way to get a job in investing is pick the most narrow part of investing and go as deep as you possibly can in it.
Because it's really hard for people to want to do this because they feel like they're limiting themselves from the optionality of, well, I don't know if I want to do private equity or hedge fund or like early stage venture, late stage venture, or if I want to be a biotech investor or whatever. All of those things sound potentially interesting because I want to get into investing.
But the problem is, When you interview for any of those jobs, you're an inch deep and a mile wide in your competency. And so you don't, you don't give people the narrowness of, Oh, well, I just talked to a guy who was thinking about biotech investing. He was really good at this. And that was actually a fortunate thing in my career is like, I early on, I was like, I want to do software.
And I want to do software. Series B investing. And I was so specific in it. And then you learn all the little loops about it. So by the time you're actually qualified to interview for it, you're so much better equipped to, uh, have a conversation with someone about it. And I think it's, it's true of startups as well, that like narrowness actually is the answer, at least initially, right.
And then gives you the opportunity to kind of build those certain concentric loops around it.
Andrew: Definitely. Yeah.
Andrew: So I think one, I think five year plans are underrated. People should have those. Two is, uh, yes, going deep matters and actually like in learning as kind of a meta skill and being good [00:26:00] at that is acquiring new, you know, knowledge about things matters. So um, maybe just to comment on each of those, uh, so first my, my dad was, uh, a big five year plan guy.
He said, sure. I don't know if you ever shared his five year plan with me, but he was always a, you know, think, think longer term. And it sounds a little contrarian to the whole, you know, kind of move fast startups, like experiments a lot. But I do think in general, people should think further out of like, well, where do you want to be in a few years?
And it, you don't necessarily have one goal. You could have like, well, there's a couple thing candidates. And, you know, you don't have to stick to them. You can change them, you know, in six months or a year. Thinking forward and then working backwards is like a real thing. Um, you know, when I, uh, got out of school, um, you know, I was a physics major and, uh, I had taken a couple of CS classes but didn't have a programming, you know, degree and want to join a software company and was like, okay, my, my kind of goal over the next few years is I want to be an amazing software engineer, uh, and, and I need to figure out what that means and get good at those skills.
And so that that was the goal, right? And then when we, when I started to think about startups, I was like, okay, well, our, you know, we're gonna build a successful startup. That's the goal. And for us, that meant, you know, profitable, you know, um, and, you know, I was gonna use the fact that I was a really good engineer as like, okay, that's gonna be my, my one kind of like, Uh, that's my niche that's going to like help power towards that.
So I think five year plans, you know, three, five year, like having a point of view on that is underrated. More people should step back and think about where do you want to be? Um, and then on like the going deep in places. Yeah. I mean, when I started the first company I worked at, um, was this big, you know, what we now call machine learning and AI, we used to call it very unsexy, like we just called data mining.
Um, so it's got, it's got a brand overhaul, but like we were doing ML and AI with these big fortune 500 banks, retailers. Okay, so there was a lot of really smart folks in the data side, but nobody knew anything about UI and UX. Well, I got [00:28:00] lucky. I happened to report to the woman that was kind of the UI person.
And this was all very early web software days, right? Versus there's no Chrome, there was barely Firefox, you know, we were selling, it was all Internet Explorer. Well, I just decided, I was like, okay, I'm going to become great at, you know, designing for the web. Well, in early 2000s, that turned out to be a great idea because that would become like the next decade's worth of stuff.
So I obsessed about that and became like the expert at the company.
Logan: we did. Did you see that trend? Or was it like, Hey, you know,
Andrew: Yeah, it felt like it. I mean, more and more things were coming out where it was, okay, you know, I mean, it's The company I was at, I think it moved away from Java applets to, you know, okay, we're going to deliver our software through a browser and it was enormously painful to develop for. Um, you know, and it's actually, and yeah, it seemed like there was a lot of momentum, you know, big companies.
Like I remember, you know, Yahoo at the time was really invested in, uh, you know, UI and how to design for the, you know, um, for the web. And I remember I would go, it's like these early, really early days, like JavaScript conferences. with like, you know, uh, it was like John Resig who created jQuery, which nobody knows who that is.
Like the founders of GitHub were there because this was like a big part of their ecosystem. You know, the folks that ran like the Facebook front end. So it's a very small, like, you know, 40, 50 people. And this is like, yeah, we're just all nerding out on this tech. So I think trying to pick those pieces of technology that you think are going to go forward, uh, is a really good, uh, idea.
And then, yeah, I mean, like you get good at that. And I think there's this, this gets this meta skill of learning. I think really good entrepreneurs, founders, you know, you gotta be, or frankly everybody, you gotta be really curious. Um, so we talk about at Klaviyo, we like people that are, uh, uh, fat T shaped.
So, you know, people talk about T shaped, Hey, I go really deep in an area and then I kind of go a little bit in everything else. We say, that's good. But ultimately, you know, if you can have like a couple of areas where you go really deep, you know, [00:30:00] You know, it's the intersection, intersection between those things, the overlap.
I mean, that's what makes you even better. Does it really acts as a multiplier? So yeah, I love this idea of go deep somewhere, become the expert at something, uh, and then start to add up a couple of those areas. And then that you can then parlay into like the next thing. And that, you know, for me, it was okay, well, how do we get over an engineer?
I'm going to get good at like, you know, building for the web. I'll also learn how to do the backend parts. Now I want to start a startup. Great. I feel like I've got the product and the engineering part locked down. And then I would just corner people about, okay, well, how do you do the sales and marketing side?
How do you drive demand? What does it mean to run a sales team? And you start to just aggregate more and more of these skills. And then you, you know, once you've kind of reached that next, so we have this saying internally inside of Klaviyo, you know, we're always 1 percent done every time you have a little bit of success.
It's like, great. Now you got to move the goalposts further out so that there's like a, you know, a bigger thing to aim at.
Logan: In, in that pursuit and, uh, uh, so, so went to go work at APT, sold a MasterCard for 600 million bucks. Were you there when they sold or was that
Andrew: No, but I'll tell you like the, so another thing is that, you know, for folks that are, you know, in their early twenties, just graduating school, Um, like I think if you've got an idea and you wanna go pursue it, great. But if you don't, I think joining great companies is a great idea that, you know, this company I worked at Applied Productive Technologies.
Logan: is, which is like the most generic, boring
Andrew: It's a mouthful. I, it, it was a very fun fact. We tried to get apt.com. It turned out like the only thing we could get, it was APT seven. The other numbers were already anyways, so very fun company. But really, super well run,
Logan: you remember how you even found it?
Andrew: Uh, yeah. I mean, so they were one of the things that I took from them is they were very big into campus recruiting.
Um, something that we've carried forward, they would just get, they were, they had a policy of, they were almost exclusively going to hire from campus and then they were going to grow folks up. That's become a little bit, that's another thing that I have a very strong perspective on is in general, you should prefer to hire folks that are either early in career or out of school.
Um, yeah. Absolutely. Absolutely. And then really build a program where they can grow up there. Uh, if you look at a lot of the [00:32:00] all time great companies, that's how they, that's how they have built culture. Um,
Logan: And why do you think that's the case? Is it just that, like, that's the, you're pulling from the broadest pool of talent and you get to refine that, uh, within your own,
Andrew: yeah, I think there's, I mean, you know, look, I mean, recruiting is hard, right? You know, uh, I think we all know that. I think everybody in their heart of hearts is like, I want to hire the person that knows how to do the thing and then I'm going to hire them and I'm just going to leave them alone and I never have to talk to them and it's great because they're just going to figure everything out themselves.
That is just practically not possible. Um, so I think if you, once you make the decision that you're like, okay, well, there's a way that we want to work and it's not inflexible. I mean, you can obviously, you could, you know, you could work as a team to redefine over time, but there's a way that we want to work, you know, and let's call that culture.
Um, I think the more that you're intentional about defining that, I think folks find there's this balance of, you know, people that bring existing skills to the table and then sort of like, you have to get them to adopt your culture versus. People have maybe relatively fewer skills, but you can teach them and they're, you know, they're good learners.
And I think you can decide which is for you. I think a lot of folks tend to think that it's easier to find the ready made, you know, kind of version of people. Um, I tend to think it's actually, it, well, it takes more time and you have to set up the infrastructure. It's a competitive advantage if you can do both.
Um, so anyways, I mean, we obviously do that a lot in, you know, disciplines like product and product. One of my favorite ones. Every time I interview somebody for a product job. So why are you interested in this? Because nobody goes and gets a degree for like, you know, product and, you know, design. So how did you get here?
And that's an example of one where we just have a curriculum internally and say, okay, if you, if you come here, I assume you're going to be say a future, you know, future founder, entrepreneur, future, you know, chief product officer. And we're going to put you through a bootcamp to get you there. Right. And it's going to be hard.
It's going to be intense, but we will give you our opinion version of what it takes to do great things. And I think if you look at a lot of great product, I mean, they've got similar things, right? For each. And [00:34:00] obviously you can then extend that to other functions.
Logan: So APT was recruiting on campus, uh, and so, so they pulled you down into that and you had, you had a nice run there, but then you were also purposeful in pursuing performable and trying to learn, I guess, as you were thinking through. Your journey at what point were you like, Hey, I'm, I want to be a founder.
And so I'm going to go round out all these skills. And then how'd you get looped in with performable to round out the sales and some of those
Andrew: So the first thing was, you know, I was cool. It's like, I just want to be a great engineer. You know, I tried to join a company where there were other people that I felt I could really learn from. And
Logan: I heard big tech, by the way, uh, like wouldn't even have a conversation with, with you when you applied.
Andrew: I mean, there's two things. Why I said I had done an internship at Microsoft, which was great, but I also, you know, I had heard these like horror stories, and this is like early two thousands of.
Hey, you want to be like on a team that has, uh, a bunch of like of good people, but not great people because we have this thing called stack ranking. And if everybody's awesome, well, there's only, I don't know, one five star one, like, you know, you're the best of the best ratings. You want to be with good people, but not great.
And I always
Logan: within Microsoft.
Andrew: And this it, it, and again, I don't know if it was a pocket team, but somebody had mentioned that that was a kind of a thing. And I was, that was like, I was like horrified by that. I was like, that sounds horrible. Like I want to be at a company where it's, it's all all stars and it's growing so fast.
Like the pie is just expanding. Right. So that's been a very key tenant of mine, like for us at Klaviyo and like all places I've chose. Um, so it was being a great engineer. By the time I left there, um, I intentionally wanted to join a startup because, you know, I was thinking I was like, you know, I had these, As I got into programming, I'd been doing a lot like nights and weekends.
Um, I'd spun up a couple of little side projects when call them companies and I really tried to monetize them. I was like, okay, this is kind of interesting. I think I may want to try this whole entrepreneurship thing. Um, I was like, well, I want to, I want to front and receive to see how it's done. So I want to join a really small company sub, you know, 10 people.
And so when [00:36:00] I, I decided I was going to stay on the East coast and it was either Boston or New York and I obviously had a bias to Boston cause that's where I was from. Uh, and so when I was back, you know, visiting family and friends, I I asked around, um, and do a little research on, like, before this, before there was Angel List.
It's like, where do you go find startups? Um, and, uh, and found this company called Performable. And I'd asked around about the CEO and the CTO, because at the time, you know, I would have been like their first or second engineer. I said, I don't even really care what they're doing. Um, I just want to work with people that are really good at the craft.
And everybody told me they're like, well, I mean, you know, David Cancel, who's a CEO and all these towards the CTO. Uh, they're like, well, they're multiple time founders and, uh, you know, David's had a bunch of like success, um, uh, at these other companies said, okay, great. Like that's, that's all I want. And so I met them, uh, and, uh, ended up joining.
I even made the mistake of, I was so into it. It was like, I don't even really care what you're building. Like I just want to work with really smart people. Uh, I also didn't care where you're working. I had assumed, you know, their Boston company is like, Oh, you must be in Cambridge or somewhere in like the kind of suburbs.
Andrew: Oh, no, like, uh, David at the time was living almost, you know, in New Hampshire. So about an hour outside the city. So I realized two weeks before I was like, Oh, crap, this is a commuting job, right? But it was totally worth it because they hired exceptionally well, learned a ton there. And we'd have these, you know, it feels a little bit like a movie, but like, we have these, you know, whiteboarding sessions, right?
Where, um, you know, the engineering, the product folks. We'll get together and you know, David, we have an idea about what you want to build and we kind of like, we'd hash it out and then literally be like, okay, cool. Like just go prototype it out, go code it. Um, and so it was just very rapid, right? You know, day over day, week over week like that and you know, you're on the front lines with customers and it was a great experience and it kind of demystified for me what it takes.
Andrew: Uh, and you got to see like, you know, how you would [00:38:00] share, like, here's the revenue numbers and here's how we're finding customers and here's why we're changing our strategy and By the time after, you know, Performable, uh, got acquired by HubSpot, um, and so, uh, and I said, I'm still very close with the entire team there.
Um, but that at that point, I was like, well, okay, now I've seen what success looks like. Like, this doesn't feel, I mean, like, there's still a lot of risk, like you can make good decisions. But now I've seen the patterns. Now I think I can really go for
Logan: How long was a performable time?
Andrew: I mean, that's great. You know, you can learn a ton in a really short period of time.
Uh, I was there for about 12 months. Yeah. And that was, yeah, it went from, I mean, it was very fast. I think they made maybe working for like six, 12 months before that. But yeah, it was very, I mean, we did accomplish a lot a
Logan: Sure. And for people that don't know, performable got acquired by HubSpot, uh, Cancel became chief product officer of HubSpot and sort of, I mean, I think Allegan gives him a lot of credit for. Uh, really rebuilding people sort of forget, but HubSpot was like, uh, sort of a feature ish initially. And David, I think had a pretty heavy hand performables business in building out into the full, you know, modular suite of products that it became today, but that was a pretty linchpin moment.
I feel like in HubSpot's journey from, um, a few products or like a discrete thing into the journey. And then David, obviously, uh, both of them went on to, uh, Start, uh, drift, which Vista bought for over a billion dollars as well. So repeat entrepreneurs,
Andrew: Yeah.
Andrew: And like, and one of the other thing, I mean, the, the other reason I recommend, uh, you know, going and finding great companies and really great people to work and learn from, um, Is, I mean, those folks become, you know, when you, when you want to start something, I mean, when we actually went and raised capital, um, after, so we, so after that, you know, we started Klaviyo, uh, we, you know, did our, you know, kind of, uh, you know, two years getting it to profitability, all this kind of stuff.
And when I went and said like, okay, I think we're ready to raise some, you know, uh, a venture round. Um, you know, I went back to David and Elias and said, well, who do you like? Because they [00:40:00] knew who all the, you know, and by the way, would you guys be willing to back us? And, you know, of course, they're like, yeah, of course.
Um, and, you know, so that was a really helpful thing because like they acted as, you know, ventures advisors. And then like, you know, that, um, you know, that's how that introduced me to Brian and Dharmesh and the HubSpot team. And, you know, we very much look at there's a lot that I think that that company actually is really well that we've been able to learn from them.
So a very big fan of this fine, great people that like know how to do it and they'll probably surround themselves with other great people. It's a great way, you know, if you're not a big networking, I'm not the big, like go to conferences kind of person. Um, it's a great way to meet the people, you know, people that can actually help and serve as you know, those, um, could be angels, could be advisors.
Logan: I want to put a pin in the, the founding story and maybe come back to that, but one of the things, how many people was Clavia today?
Andrew: Uh, we're just about 2000.
Logan: I know that's a vanity metric that, uh,
Andrew: We have two vanity measures. We don't celebrate. We don't celebrate fundraising. We do not celebrate how many
Logan: yeah, but it's still, it's, uh, it's, it's a lot of people within the organization and, uh, along the way, I know, um, your role, like anything that happens with a, uh, CEO growing, uh, as the company grows, you're. Your rate of learning and development and skills need to outpace that by which the company is growing.
And as you guys have grown really quickly, uh, your skills have had to grow very quickly as well. I guess, um, there's been a number of different functional responsibilities that you've probably had to see it along the way. And I've, I've heard you use the, uh, the phrase giving up your Lego blocks or whatever it is to describe
Andrew: a great talk. Somebody does about like, right. Giving up your toys, which is totally the right mental model for like, you know, growing as a leader.
Logan: What did you hold onto, uh, longer than you should have in that journey?
Andrew: Gosh, probably too many things. I mean, like, well, the earliest from the early days is something I would, you know, we talked about how like Uh, constraint breeds creativity, and I think that's, I think it's very true. Something I think I would have done differently is we probably waited too long to hire, you know, other engineers.
Um, you know, for the first few years I was, you know, owned all the code, and I was like, okay, well, I've kind of, I figured out a [00:42:00] way to make this work, you know, by myself. Man, there was such a massive unlock once we actually started to grow that team. Um, and I think that was, you know, I think that's usually been, you know, I, I generally would say I'm very slow to hire.
Um, and you know, I think there's, there's this norm in, uh, startups, you know, uh, I know this is entirely intent of like, you know, re Hopkins book, blitz scaling, but there's this concept, you know, this concept of like, just go fast, right? Like the idea is speed, go win the market, um, because you know, there's sort of first mover advantage.
have often says like, No, I'd rather not do that. Like, I'll be slow. And I think that's generally worked well for us. But I think my biggest mistakes, like the things I've held on to are things that yeah, we either I should have delegated that to somebody let them run with it. And, uh, or we should have just grown a certain team faster.
So, uh, I mean, I think for years we were a very lean, um, you know, uh, product engineering group, which is good because it, I mean, you get to be able to really understand everything. Um, but in retrospect, it's like, boy, we could have gone a lot faster. Um, yeah, after that, I mean, my, my, this isn't quite like which things to give up, but I, uh, somebody wants is probably four or five years ago, gave me the advice.
They're like, look, Andrew, there's, there's effectively three jobs you can have. One is when you're the person doing the work. Okay. So we think of like the, you know, the corporate IC, right? You're the expert. You're getting the work done. Okay. So that's level one. Level two is now you're like the leader of a group of people that's doing that, right?
So you're the, you know, the manager, um, and that's, now there's one level of indirection. And then the third is like, okay, now you're the, you know, you're the, there's one more level of indirection, right? Um, you're the like, lead of lead leader, leader of a bunch of leaders who themselves have teams. I think that's actually a really good mental model to think about, like, what gets challenging?
Because for a lot of new, uh, leaders, I think the challenge is like, well, gosh, I was really good at this. Like, how do I coach and [00:44:00] mentor and evaluate and like, gosh, I have to set the goals, but make sure I don't do the work. Right. And I think that's a, that's a, that's a skill. Um, and then there's this like level of indirection where it's like, okay, gosh, well, how do I, you know, influence and inspire a really large group where I may not know what's going on or, or what's my, what's my operating system such that I can inspect what's going on several layers down the stack quickly.
Um, when there's more complexity that I think developing that, you know, kind of like that tier three model, um, is something that. If you realize that's coming, sort of getting in front of that and starting to build that faster will help, um, you know, feedback that I've gotten over the years is, uh, you know, the clear, you know, that I make my operating system like the way the interfaces that I'm going to have the way that I want to work, um, to others, you know, the easier it is for folks to go execute.
And when there's not clarity on that, not just like what the strategy is, but literally like, okay, so when are we going to look at that and review that and who makes decisions? That's often where things slow down the most. And I've often felt like, okay, well, I feel like if I tell you what I'm going to do, like, that's like constraining you too much, but actually like, no, no, we actually want that constraints.
And then we'll operate whatever with within whatever bounds it is. So for instance, our product team, um, you know, we just have a norm of. Hey, here's how we're going to evaluate which project we're gonna work on, what our strategy is gonna be, what goes on the roadmap. And it's very clear about like, Hey, these are the check in points, right?
Um, and that actually gives a lot of freedom for folks to experiment with ideas, uh, you know, and off they go. And then we have another process by which it's like, if you want to spin up a whole new product inside of Klaviyo, great. The doors are always open. Um, but these are the requirements, right? Here's the sweat equity you got to put in, here's the prototype you've got to bring.
You got to have some idea of the projections of like, what is usage or revenue look like? 12, 24 months, not any different really from like, you know, pitching, right? If you're going to go raise outside capital. [00:46:00] But the fact that we have that has been an enormous accelerant. So it's a lot of like setting up those structures, you know, lightweight enough, but deterministic enough that folks can then, you know, operate within that.
Logan: So there's, there's, I guess, two components of that, I think, in the higher abstraction of, of being a manager of managers. And one of them is the operating principles and some of those gating milestones. And I think the other one is, uh, probably some culture tenants related and the two aren't totally, uh, uh, disparate from one another.
But I'm curious, I, when going back and prepping for this, I was listening to some of the older podcasts you had done. And one of the things that you. You mentioned that you wanted to get better at, or one of the learning things that you had maybe talked about four years ago was actually setting up a culture of, uh, it's not experimentation.
Cause I think you guys have probably always had a culture of experimentation, but product development or founding development with it for new business units and all that within the, within the business. How have you, it sounds like you've made progress on it, given what you just said, but how have you sort of set up those frameworks and milestones and like the discrete points of the journey along the way?
Andrew: Yeah.
Andrew: So I think there's, we have this concept internally of like, you want, you want a culture to have high entrepreneurial density. What does that mean? It doesn't necessarily mean somebody has to have gone and founded a business, you know, Um, but you want folks that will take ownership and sort of say, Hey, if there's a problem, I'm going to go solve it.
So okay, if you've got, say, then, uh, you know, high density of folks that want to do that, you know, a lot of them are going to come, they're going to bring, they're going to bring problems or solutions and say, Hey, can we just go do this? How do I fund this? So yeah, well, we, we're not, we're not perfect at this.
Yeah. We've done a better job on the product side than we have in other places like, um, but yeah. Uh, But yeah, you want to say it's like, okay, cool. Well, cause the question everybody comes back was like, how do I get this funded? Right. Cause I have this idea and I want to go do it. Like how do I make it my full time job and how do I get other people to come with me and work on it?
So, uh, it's interesting. We've taken, I mean, really just off of, uh, the way that, you know, uh, venture capitalists or investors make decisions. We've [00:48:00] taken the same kind of like seed series ABC model and just set that up internally. Um, so I was actually doing this a few months ago is, uh, you know, we, we do check ins.
We have a sort of a different cadence than we do with our regular, like kind of core product roadmap. In those cases, I treat those very much like, yeah, you're a startup. So that's, I mean, you know, when we were the first, I don't know, three, four, five, six years, we didn't have board meetings. We would just check in, you know, every six, eight weeks and there was kind of like a standard little format we'd have of like, okay, you know, like what are our, what are our core metrics of our business?
And even to those folks, we'll expose to them like their P and L we'll say, it's like, look, this is how many people are working on this project. This is how much you cost. This is how much revenue you're bringing in. And what was interesting is we, I don't think it was a really, we found that people are very motivated by that when they can see that and they can see their burn rate and they know, okay, gosh, here's how I'm trending.
Right. It, I think it even more forces this like ownership of an urgency around what they're trying to accomplish. Um, so we've done that and we now have, uh, gosh, probably like half a dozen kind of like at scale, like, you know, like there's, there's a meaningful number of people, like not hundreds, but like, you know, tens of folks work on things.
And uh, since we've started that a couple years ago and uh, yeah, I mean, I think that's we talked about like, you know, uh, evolving into offering more products. You know, for instance, we believe Kladio on top of um, you know, this database that we built, like marketing is just the first interface. I mean, there's plenty of other interfaces between a business and a consumer.
You know, like what about customer service? Like what about? The actual, like the website or the mobile experience, like there's plenty of these things. And we have people that are working on all of those, right? Um, and we've followed this model and it's interesting, we've tried to resist hard this urge to kind of centrally plan those.
So it's a discussion debate. We have a lot of like, when does it make sense to kind of centrally plan them? Because we don't have to move quickly versus when is it good to have this kind of like startup model? And I think our general philosophy is it's better to aim for the, you know, kind of startup model where we can, if it's really [00:50:00] urgent, then I think we should have urgency in finding the right, you know, founder, so to speak for that business, um, you know, to kind of drive it.
Logan: In terms of giving autonomy, uh, to the people, the founder within it, uh, versus Um, the, the, uh, considerations that lead to the, the shared service on the back end. So it's not discrete products. Like I assume you're not allowing people to be picked between the cloud providers or the databases that they want.
Like how do you sort of think about where you're going to make it autonomous and empowered versus uh, shared
Andrew: That's a great question. I mean, it's, it's something that we've, we're, we've, we've learned a lot, a lot of battle scars in that over the last few years. So I give a lot of credit to the like, You know, to the, the Clavios that like kind of were on the first teams that did this
Logan: Clavios is
Andrew: is our employees. Yeah, yeah, yeah, our, our, our, our, our employees.
Uh, yeah. So for instance, actually the first, the first team that did this actually set up a whole nother cloud hosting accounts because you know, they branded it differently and everything, right. Cause they want to be totally different. Um, I think they found actually that it was easy. There's a lot of like, I mean, there was some freedom from that and some things that were like, Oh gosh, this is redundant.
So. Yeah, I mean, look, look, when it comes to there's primitives that we want to offer, you know, we're, I think our customers expect of products that come from us. So, uh, things that are secure, things that are reliable, things that will scale. And so those are primitives that we have a platform team that just, yep, here's what you get for building blocks.
I think a lot of what we learned was, well, what are those core building blocks? And what are the interfaces that teams need to go be six to be successful so they can have autonomy? Um, that's probably was the biggest one. There's, you know, there were some other things of, you know, hiring was very interesting.
Well, do you have to follow the same hiring process or not? Um, and so we've largely said, yes, we should have some norms around hiring. But like, you know, you can make the decisions, uh, you know, somewhat, somewhat more autonomously than maybe our core core teams do. So [00:52:00] there's a lot of little things like that that we've had to sort out.
Um, but I think once you do, I mean, we're really trying to. I've tried to put people in the, because I, there's, there's no doubt in my mind, I would have potentially stayed if a program like this had existed at the first company I worked at, um, ABT. And, you know, it's no fault of theirs, but like, I think back to this, if there had been a program like that, where they said, Hey, we're really going to put you out on an island here.
You've got to sink or swim all by yourself, but you're going to do in the confines of this organization. It's probably a good chance. I would have stayed for a lot longer. Um, and I just, it wasn't really something that they were anticipating or considering, but it's, it's definitely made me think about Klaviyo because I mean, we've had now dozens of people go off and start companies outside of Klaviyo.
Um, which is great. Um, it's one of these like kind of very bittersweet conversations when somebody
Logan: Sure.
Andrew: I'm sorry, but I've been on this thing and I'm leaving and it's like, okay, Godspeed. Like, uh, how about it? I think a lot more of those folks, you know, a bunch of folks that are very real will say, well, actually, maybe I don't want to strike it on my own.
I actually like this, you know, product category. I just need a construct that gives me more of this like freedom because I do want more risk reward.
Logan: Yeah. It's a double edged sword of recruiting the entrepreneurial, uh, person that, uh, that, that might leave the confines of the organization at some point.
Logan: It sounds like you guys, um, if, if there's two things in hearing, uh, you talk here and then also in doing prep for this that you maybe uniquely saw for is, and I, I'd love to, um, talk about how you, because you alluded to hiring, How you tease this out in interviews.
One is kind of this individualistic, um, empowerment, entrepreneurial side of things and figuring that out. Most people I've found, it's much more comfortable to be told what to do than to stare at a blank piece of paper and figure out what to do. It's just, I think 95%, 99%, 90%. I don't know what the right number is, but are people that, uh, I'll colloquial call it.
football players versus, uh, basketball players. Like football players, they get told what to do. [00:54:00] They do their job and they do it well. They can be great team members, but they're not creating in space the way like a basketball player, a soccer player will do. I think Keith or boy has something barrels and ammunition, uh, the people that are firing versus the people that are aiming and all that.
Um, how do you go about? teasing out the people that are individualistic enough, empowered enough in an interview process. Because I've always just found it's a, it's a very hard, everyone says the right lip service of like, Oh yeah, I'm a self starter and all that. But how do you guys go about actually finding that?
Andrew: Yeah, it's a good question. I don't think you can kind of manufacture a, you know, a question or a case study and say, here, kind of solve this. Prove to me. It's all a track record thing. Um, so, uh, there's some skills that I think, you know, you can only show by having done them. And, uh, you can end up in these spots where somebody's like, well, no, I'm telling you, I really am.
And I'm like, well, if you were, you would have done it already. Um, so our trick for that is, uh, I, or our technique is I'm a very big fan of like kind of, you know, the resume walk or kind of the life story. I think a lot of people, when you boil it down, like if you tell us, tell me about the projects you've worked on and if you ask kind of really kind of like detailed questions about like, well, how did you make that decision or why did you decide to do that?
Um, you know, entrepreneurial folks, I find like they won't go to others to, you know, ask, well, what should I do in this situation? And they'll, they'll have a higher percentage of times where they're like, Well, I wasn't sure what to do here. And so here's how I came up with a strategy. Here's the research I did.
And here's the, here's how I thought about the options. Here's my experiments with didn't work. So I went back and did this other thing. They have lots of stories like that. And I think one of the things that we're careful about is it doesn't, you know, entrepreneurship does not equate to like, you have to run a business.
It can, it can, I mean, there's, it's everywhere. And so it's just sort of asking those questions. And the fun part is, I think most entrepreneurs, I mean, look, they kind of wear it as a badge of honor of, you know, well, yeah, let me tell you all the things I tried and this one failed. And like, but then I succeeded.
I mean, they almost can't [00:56:00] help but talk about it. So we love that kind of, you know, a little bit life story. Um, and we look for those folks that are kind of proactively decisive. Um, and if they're doing that, then I think there's a good chance that when you put them in those spots in the future, Hey, you know, They'll, you know, they'll, they'll sort of surprise you with what, with what they've done.
Um, but it is one of those things. I mean, there's a couple of skills like this, like, you know, that's one of them where, you know, our advice to folks is like, you kind of can't, the only way you can prove to somebody you do it is by doing it. You have to have a track record. You know, another one that we, uh, I often look for, um, you know, is folks that are, what we look for all the time is like folks that are like super curious.
They're really good learners. And it's one of those things where it's, it's actually really easy to vet. You just say like, that's awesome. I'm sure you're learning about something right now. What are you nerding out on? And you get sometimes people that give you the answer of like, well, for my job, like I have to read this page.
I'm like, hang on. Like that sounds like a everyday thing. And then you get the people that are just naturally, well, yeah, I'm reading this book and I think that's really interesting or I listen to this podcast or I watched this video the other day. And sometimes it has nothing to do with their day to day.
So it's just something that they're passionate about that they got into. And I think what's hard is we all, you want people that are like that. And oftentimes, you know, I think you hear people, it's like, well, I don't, well, they don't, they don't bring it up. You're like, well, that's okay. Maybe I just asked the question the wrong way.
It's like, no, like the folks that are truly curious and are learners or entrepreneurial, it'll be so self evidence. The real challenge is, I think, just, you know, showing that is finding those folks. So one, and in our, one of our, you know, one of our strategies for that is. We just advertise straight up, uh, you know, on our website.
Like when I go talk to folks, I'm like, look, I'll tell you the cheat code to get into our company. Right. But you have to have, you have to, you have to, you have to be able to tell me you have a track record of these things. So if you don't like rates, like tonight is a good night to start working on this.
Um, and if you don't want to, that's fine. But like, basically we'll tell you what the test is going to be, so to [00:58:00] speak, what our bar is. And then you just need to bring the track record. Um, and I think actually advertising that we're for that. There's a lot of people that are like, you know, gracious learners or there's actually, I think those people like to cluster around each other.
And so if you just kind of advertise it like, hey, we're gonna be this beacon of that, then actually you'll get a lot of people are like, oh, that's good because I was looking for that. And a lot of companies, I mean, a lot of companies frankly don't want people to be entrepreneurial. Hey, we've got the plan.
We kind of need you to do your job. As long as you do your job, you get rewarded. So I think by us saying like, no, we actually, we, we want, we want you that you're a little rebellious and you get think outside the box. It's actually you can get people to attract themselves and you know, their friends.
Logan: Hmm.
Logan: You believe that resume should be laid out in chronological fashion rather than reverse chronological fashion. And we talked a little
Andrew: oh, yeah, yeah, yeah,
Logan: the narrative of the journey and hearing about the inputs and all. Can you just I think it's an interesting point or it's something that I hadn't thought of because normally resumes.
Okay, here's where I was. And then here's where I was before that. Here's where I was before that. And then your eyes sort of drift. And by the end, you've lost the
Andrew: Nobody really looks at the first stuff,
Logan: Yeah, exactly. By the end, you're like, okay, you did that seven years ago and let's talk about this thing. But your eyes gravitate to what's top of the page.
So, so it's, it's an interesting point. Why, why do you think that, uh, laying it out in that way is more,
Andrew: I mean, look, I mean, like, obviously somebody's like, recent experience may be the most relevant to this thing. Um, I just love asking folks, you know, I love asking folks like, okay, cool. So, uh, you know. Tell me about like after school, you got this, you got your first job, uh, like Tony, why did you decide to do that?
Um, you know, what, why'd you pick that company and you know, what'd you accomplish there? And then I find that when you do it in this kind of, you know, chronological way, uh, folks actually open up a lot more. They're like, Oh, well, let me go back in time. Yeah, the reason was this. And it's like, okay, now think about that stretch, right?
Say it's a couple of years. What are you most proud of from there? Oh, okay. Well, that project was really interesting. What'd [01:00:00] you learn? And so I feel like it's actually easier to kind of unpack for, you know, um, so I'll often guide folks when I meet them. I'm like, Hey, look, I'd love to talk through your kind of story, but like, let's, let's do it from the start.
It's, it's actually very disarming. A lot of people are like, Oh, you mean like where I got my first job there? I mean, especially somebody has like a, you know, say 10, 20 year credit. Oh man, it's so long ago. I mean, you know, you know, it, it really takes the pressure off and feels less like an interview. Um, I mean, there's obviously merits to like the things that are maybe most relevant, you know, are, you know, more local, but I take that as part of the reason I believe that when you're actually trying to understand somebody, you have to look at their whole body of work.
And if you're just looking at the last, you know, say the last thing they did the last year or two. It's very transactional. It's just like, look, I need, I need somebody that does this job. Can you just tell me, have you done this job recently? And if you have grades and it's like, well, I mean, you know, I'll say somebody is, uh, you know, uh, 30 or 40 years old.
I mean, you're literally looking at like three, you know, like, you know, three or 4 percent of their life, right? And it's like, that's not a very, you can just, people are actually willing to share more than that. And if you just, you know, give them a chance.
Logan: there's a derivative consideration of it, which I thought maybe you're going to say and maybe is implicit in this is the rate of learning and the inputs into the deciding of each step along the way, like I, people would jump jobs and that's fine and you can switch, but like. Hearing the input and the process by which they, they did that, did they learn new things in that job that led them to taking the next chance along the way?
And what were the inputs into it? What did they hope to accomplish? What went right or wrong along the journey? And then what did they then leverage from there? And generally it doesn't need to be a linear path, right? It can be step functions along the way in a non linear one. But it is, Very informative to talk through each discrete decision and you don't get that in the reverse looking way of, of, you know, what you've done most
Andrew: Yeah. Definitely. Yeah. I mean, first of all, I'm a big [01:02:00] advocate for him. Both my parents have had, I mean, I think they've had like one or two jobs, their entire careers. And I think that's, I think that's a little crazy. I mean, it's just, but it's certainly now it's a lot easier, uh, to, I think you should try different things.
Um, there's a Richard Feynman, uh, physicist, I care about, I always get this wrong, but he went to undergrad on either the East coast or West coast and then switched for his grad work. And it was the fact that he did both that he understood. He's like two great departments. I think it was like, I think it was Berkeley and Princeton were the two.
And he learned, he's like, there were very different cultures, ways of working, but like it was actually doing both that like made him who he was. So there's nothing wrong with like, I mean, you should, I think you should have a velocity different projects, whether it's been inside a company or just try different companies, but it's, that's exactly right.
It's this, what is your logic for why you made these decisions? And, you know, it doesn't have to be a master plan and the master plan doesn't have to like necessarily work out the way you thought it would. But having a point of view on, you know, Oh, I was intentional about going from this to that because of, you know, for this reason I wanted to learn or try this thing is a massive tell.
And like, yeah, I mean, all the, all the great people, when you ask them about that, I mean, they're happy to tell you about this cause they're like, yeah, I was thinking through this and maybe this, this works. I learned I didn't like that, but I really liked this.
Logan: Because if you're not purposeful or considerate about your own career, which Maybe outside of who you marry or whatever, a few other considerations. It's probably on the discrete project I give you. You're probably not going to be like if you didn't choose in a purposeful manner around that. Have you ever read the book range?
Uh, it's, it's interesting. I mean, uh, what you said about East Coast and West Coast and, and pulling from different threads together to, to, to. to solve problems. It's very, um, it's, it's, uh, it just reminds me of, of that, of how, how the, some of the breakthroughs we have as society come from broad experiences and pulling all that stuff
Andrew: Oh yeah. Yeah.
Logan: Yeah, there's a number of things. I mean, I found in some ways, the, the, uh, venture [01:04:00] capital job itself actually lends itself to two of the character, both those characteristics curious for obvious reasons, you're going to look at a bunch of different things, but then also. Um, there's you can justify your day is going to start blank for the most part, like, and you can do any, you could have a podcast, you can, uh, whatever, go to networking events all day.
You can fly to a conference like you can justify all these different things. And most people are actually pretty, um, you know, It's, uh, it's discomforting to not really be told what to do in general and how to spend your time, especially when there's a whole bunch of mental models of how people have been successful in the venture capital industry.
But I found it's interesting. You could find people that are entrepreneurial in any field. thing that you can just ask the question of, like, what in your job was kind of broken or at your school was kind of broken? And what did you create to fix that? Right? And maybe they started a club, right? Or maybe they, uh, redid the comps.
They're an investment banker, and they The way they were doing comps didn't make sense and they redid it from the scratch or maybe the recruiting process was broken and they just took over the high school or the junior or whatever, the college recruiting process or something. Like, I found that all of those little things, it's really hard to fake that stuff that you just saw something where there's smoke, you, you, you yelled fire and you went Fixed it right and like that manifests itself no matter what the job is or even if it's not a job if you're in college or those types of people tend to find a lot of different stoves to touch and fix those problems and so teasing it out is always a little hard but generally they go
Andrew: the good news is, like I said, is I think most people don't like, I mean, they like to gush about those things. They won't hide them. So if people have those stories, that's right. And you kind of pro offer it up. Hey, give me a time when something was broken, you fixed it. I mean, folks that have those are like, well, let me tell you, right?
So it's, yeah, it just, you just kind of have to ask and you have to be, you know, Look, I mean, not everybody, like you said, not everybody operates that way. You just have to be willing to, that, Hey, if somebody doesn't, you know, if somebody doesn't give you [01:06:00] any evidence of that, it's not probably because you asked a bad question.
It's probably just because maybe they haven't done it.
Logan: Which is, which is fine. And there's a lot of great companies that people can go work at, but it's not going to be the same individualistic and empowering, um, of, of that now. Um, so operating principles and, uh, like different things you guys espouse internally.
Logan: And we've been, we've been talking about some of them related to hiring and maybe I'll circle back to hiring in a second, but, uh, eating your vegetables, uh, is, is an adage, I guess, or, or something of a core principle that you guys, uh, speak to.
Can you elaborate on what that is? Uh, when I heard it referenced, I didn't. Here are the full depth of where it comes from and what the purpose of it is. So I'm curious.
Andrew: little bit from when Ed and I were, there's just the two of us. I think a lot of it's, you know, it's, it's always kind of sexy to work on, you know, the flashy, you know, it's, it's the net new, uh, project products open up a new market. Um, you know, we, we talk about like fixing the kind of the, whether you call them paper cuts or the loose nails, like, but going and doing that work, which is entirely unsexy.
That's where this eat your vegetables, uh, kind of comes from is, yeah, that's actually celebrated inside of, uh, so we have a whole thing, this kind of ever evergreen, you know, I guess theme initiative inside the company, which is like, how do we just automate more stuff away? yeah, that's it. And just like teams that go around and are just like, just constantly finding problems where it's like, well, that, that seems silly.
And like, let's just audit it. And you know, we have doing all hands every week. And I love when we celebrate things like, Hey, here's this thing that's maybe it was like an employee process or maybe it was something for our customers or whatever. It's just something that's that just sucked. And it's like, Oh, yeah, that used to cost, I don't know, it used to cost me an hour every time they did it.
And now it costs a minute, or now it's actually just zero. And it's just celebrating that. And so the kind of eat your vegetables is a little bit like, Hey, we're gonna celebrate that. But also, the best time to solve those problems, if you're scaling is today. So this comes up all the time, you know, it's, you know, I'll use an [01:08:00] engineering analogy, but like, it's everywhere.
Transcribed Is if you have some processor system that's not scaling, and it's, you know, kind of on fire, and I remember in the early days, like she built some system and, you know, it would break every now and then. And it's like every time, you know, you two acts would break twice as often. And at some point, you're like, Well, this is silly.
We need to fix this. And we've always had this mentality. It's like, if you just want to go solve and eliminate those distractions. So, you know, before you like, eat the dessert or eat your make, just go do the stuff that feels like it's going to stink, because it's going to free you up to do a lot more. So I really, I love that.
And then there's maybe a side thing for us, which is it's not quite the same, but it's, we have a very humble culture. Um, uh, and I think there's, you know, you want to have enough confidence that like, Hey, you know, you know, you're, you know, you're going to win. You've got a strategy to really believe in.
Right. Um, But we have the humility to say that it's like, look, the unsexy stuff like is not, you know, beneath us, right? It's not somebody else's job, right? Like I get excited about that stuff. Um, and in general, this kind of like, we're going to roll up our sleeves to do the things that we know are going to long term, you know, give us the most leverage.
be most impactful. I think that's a big part of just how we, um, the kinds of people we, uh, you know, like to have at Klaviyo and then how we like to work.
Logan: The companies I have that have the most, um, I would say entrepreneurial, uh, employees within the organization, uh, that, that, that spin out the most future companies and all that. that there's a risk or a potential tension that, um, that taking things from zero to 90 percent done is a lot more fun than 90 to rounding out all those edges, uh, at the end.
And so do you hire for different people that enjoy chewing the glass of the final 10 percent or is it just simply the celebrating of it that gets the people that are entrepreneurial to go do it?
Andrew: We try to avoid this like, you know, hey, there's folks who [01:10:00] get to do the fun part and then the other part that have to chew glass at the end. Um, yeah, I think a lot of it boils down to, sorry, so for example, like, uh, one of my favorite things to do is when we're nearing the end of a project, um, you know, I make this analogy of like, if you're, you know, you say you build a new house, Uh, and you know, you're, you're the master builder and you're walking through and you're like, okay, well there's like a loose nail there and crap, this faucet's not, you know, doesn't work quite right.
And you make this big punch list of all the stuff that you can do, but they're all like easy things. So one of the things we do, because when you get towards the end of a project, you're like, oh man, what do we have to do to ship a ship? We have this moment where we like to just build that, like, you know, that house walk through and we give everybody, it's like, okay, cool.
Just go through, like, let's find all the loose nails. And like, let's, we're going to bang down all the ones we can, anything that's not, we're going to put in like version two, right? And we're just going to ship this thing. So I think that that putting that in there and then kind of celebrating that as a moment in, uh, and then having that, that ship date makes a big difference.
I'm also very big fans of You just got a ship. I mean, there's nothing like the, you know, endorphin rush of thing is out the door and it's not just a product, right? It could be, um, you know, hey, we're working with a customer on a project and we want to give us advice. Like, let's just get that out the door.
So, uh, one of the things that I think is it feels a little paradoxical, but the folks that do the highest quality work, either individuals or teams, Also, are the ones that ship most frequently. Uh, it's something we talk a lot about, like there's a thing like, oh, no, no, you know, because you see sometimes the outside, you're like, well, they do really high quality work, but it only comes out every now and then.
And so boy, they must just take a long time on every decision. And it's like, no, I don't think that's right. It's just they're iterating in private. You can't see it, but they're really moving fast. And so that's the thing that we try to, you know, coach folks is, Hey, look, if you want to do really high quality work, the best way to do it is lots of reps and sets.
Like you're going to make mistakes and we can do that in private. But the idea is, you know, like cycle time matters. So, hey, if we're, you know, uh, you know, we have a big writing culture inside of Klaviyo. If you're going to write up a narrative, uh, or a [01:12:00] memo on something great, it's like, well, I'll give you the draft in two weeks, two weeks, give me, give me the outline, you know, tomorrow.
Right. And let's iterate on three, four times, and or you should iterate on three, four times. And that's how we get things better. And then this really pushing cycle time. I think makes makes a difference. Um, you know, I think it was, I can't remember who said this, but, uh, but this idea of, you know, companies all have their kind of internal clocks and they can totally be tuned, but there's a difference between companies that are like, okay, I'll get back to you next week versus I'll get back to you tomorrow versus I'll get back to you later today.
And I think we would all bet on, it's like, oh, okay. The team that has like, just the faster cycle time, Like, again, they may not get it right the first time, but they're probably getting more shots on goal and ultimately be more
Logan: yeah, it's a trying to do this within a venture firm, which is a weird organization in general to do it, but. I'll say, um, like schedule a meeting, uh, particularly soon and, uh, well, that's too soon. I was like, well, we we have to have the first meeting to have the second meeting because the first meeting is like, okay, none of this stuff is gonna be totally, uh, fleshed out.
Um, the learning thing we've talked a little bit about the, uh, the founder experience and the, the, the journey that you guys are, um, you want to impart to the different employees along the way to set them up to, I've heard you say, I mean, we've talked a little bit about this, but like you philosophically are training people to be future founders, even if not outside the four walls of Klaviyo.
Like you, you want to build that skillset.
Logan: Um, Klaviyo University is a thing that you guys have started. Can you talk a little bit about that? And like, what are the. System was systematized, uh, elements of like actually training people to be future founders. How do you think about it?
Andrew: And look, I, you know, you, you got this progression or at least I have of, you know, what does success look like? And again, you kind of can just find it, but this goes back to like, you know, a five year plan is a good thing. Um, you [01:14:00] know, when we started, it was like, okay, let's make sure clear is a successful company.
Like it's viable. And then, you know, you start to say like, okay, well, great. Well, how, how much impact can we have? How many customers can we have? I mean, we, we definitely think we can have millions of customers. Okay, how core can we be to, you know, how they run their business, how they evolve their business?
Oh man, we can do a lot more than we're doing to say, she started to build a bigger business. And then at some point, I mean, I've had this experience now a couple of times with working with other folks that built great companies, but then they have their kind of tree. of folks that, you know, they had an impact on.
And, uh, you know, we very much look at, like, how do we pay that forward? So I think part of that is the world just literally needs more people creating, um, creating companies, uh, creating businesses, more entrepreneurs. And so we want to demystify that as much as possible. So that's part of it. We're also very happy is, you know, we've had folks go off and become leaders at other companies, and that's great.
So I think, you know, a lot of companies get admired. Because I have such a great leadership development program. So whatever those skills are that matter, right, they're like, Oh, um, folks go off, they start new businesses. That's great. They must be really creative. Or man, those are great operators, leaders of businesses.
So we aspire to, you know, have a lot of folks go do both. So collegiate university, uh, is something we saw a concept we started a while back, which was, you know, built this idea of, okay, well, if you're gonna hire people that want to learn a lot. I mean, one of the things that actually happens pretty quickly is folks start to say, Oh, Well, what should I learn?
Like help me aim.
Andrew: Um, you know, one of the first things we have is policy inside of Klaviyo, it's our free book policy. Um, one of the things David Cancel gave me for a performable that I was like, Oh, it's a great idea. I'm going to definitely going to borrow this because he said, Andrew, nobody has ever gone broke buying books.
So if you are a, if you are a learner, like you should just read as much as you can. And I think one of the things he suggested, but we did, he might've done a set drift. We then pulled into Klaviyo was like, okay, we're gonna make it really easy. If you want to go learn about something, like you can go order that book.
You know, I don't even want you to worry about the fact that cost 10 bucks, right? Um, now, uh, the [01:16:00] secondary effect of that was almost instantly go like, well, this is great, but like, what's our reading list? You know, what books do you recommend? And so we had to build out this whole library of like, okay, well, if you're trying to learn about this, like these are good biographies or autobiographies of founders.
These are like the business books that we think are actually worth the time, all this kind of stuff.
Andrew: Um, so, uh, that became kind of the start of Clever University and said, okay, well, how can we form, uh, formalize this even further? And I always liked the idea of, you know, in college or, you know, in, in school, uh, there's always like kind of a core curriculum that you have to do.
And then there's like a bunch of electives. And you know, I looked at those courses as I'm like, okay, well, it's like, Oh, what does somebody do? Somebody just basically curated, like, look, if you do the following exercises or the following stuff, like you'll gain this skill. It's like, okay, well, we should try to systematize that rather than just expect people to go figure it out themselves.
Like, let's, you know, let's figure out how we do that. We can build the content ourselves. There's so many, you know, element like, you know, systems or content online. We can pick out and curate what's best. We've had folks come in from the outside to effectively be like professors inside of our university.
Andrew: And so what we started out with was, well, what's the core curriculum of Klaviyo? And, uh, it started actually a bunch of years ago, we said, like, okay, well, let's get you two classes you must take. The first is you have to know our customers, right? I think every company should have something like it. So we literally would say it's like, okay, everybody has to spend two days a year.
And we're going to haul a bunch of our customers in or invite them in, and they're gonna do a panel and they're going to give you all the nitty gritties of like how their business actually works. And most of it has nothing to do with our software. So, because we work with a lot of retailers, e commerce, we literally would ask them, how do you go about sourcing products?
How do you do manufacturing? How do you do design? How do you decide what the next product is? Fascinating conversations. It kind of gets you to be flying along like, you know, when you talk about these, like, these iconic businesses, uh, you know, whether they're like, you know, doing, uh, clothing and apparel or toys or other electronic, how [01:18:00] do you figure this stuff out?
So you learn about that process. Then you think about how do they do distribution and you start to realize, Oh my gosh, a lot of these businesses are so dependent or reliant on, you know, uh, wholesaling on these distribution networks and a big part of their strategy is to shift that towards more. Well, we want to go direct to the customer.
We want to at least have that relationship with them so that we can tell them about what's new and, you know, and personalize that experience. And so then people can start to tie it back to, Oh, okay, God, I kind of see where we saw it. So that became class, you know, number one, right? And we literally print this little like course catalog book.
Then we started thinking like, what are other things we just want to level people up on? Another one that was kind of obvious was like writing, gosh, a lot of people go to school and they write and maybe take one class in college and they never write again. And I've seen this thing where folks join, you know, I don't know, like the business world and all of a sudden they're writing just like, They don't write like they'd write to their, their mom, their dad, their friends, they write in this like business speed.
And it's like, where did that come from? They're like, hang on, we can fix this, you know, like rather than and rather than coach it one by one, it's like very painful. It's like, let's just, let's just go back to like, you know, it's not quite how to write a five paragraph essay, but something like that, right?
So writing became a thing. And we just said like, look, there's gonna be a couple of these core classes. Part of the contract of working at Klaviyo is you're invested in learning, we're gonna help level you up on skills that are gonna be useful here or somewhere else. But, you know, our commitment to you is we'll, we'll set the table, but you have to go take them.
Right. And this is like, you know, it's, it's a little bit outside of, you know, the normal, like, uh, uh, office hours, all this kind of stuff. And uh, so we started with that, uh, and that went really well. And then we expanded that to say, Hey, we've got some kind of electives. So you know, this was, you know, four or five years ago, we had a lot of folks who were really interested in like machine learning.
Hey, I just want to get deeper into that. A lot of folks in the engineering team, but also like folks on our ops team. So we're just like, I've got all this data. I know I should be doing more, but I, how do I do this? We're like, well, look, it's more fun to do it together. So we, in that case, we just, we [01:20:00] hired somebody to come in and we're like, look, I want you to teach a class for three months.
We'll meet every Tuesday, Wednesday evening. And uh, off you go. So we're very big fans of this idea of, you know, great companies. have these kinds of behind the scenes, like training, you know, uh, uh, you know, uh, universities or like, you know, kind of curriculums. So we should build out our own. And it's the kind of thing, it's not, it's not very sexy, like something you talk about all the time.
But I think it's one of the things that how you could build up some of this entrepreneurial talents, uh, outside of just like the
Logan: yeah, how often do these things run? Like what is, it sounds like the customer thing everyone does. Is that a part of the onboarding process? Like the writing thing sounds like everyone does. Is that a part of like, when did these fit in? How has it been systematized?
Andrew: that's a good question.
Andrew: So we've actually, I mean, uh, we, we were really good at this, uh, right up until COVID. It got a little harder when everybody was, uh, went remote. So now, yes, everybody meets customers, learns customers as part of onboarding. Um, and we're actually right now in the process of, uh, like revamping that now that we're going back more into offices where we can say, Hey, okay, now we can pick the days.
Um, and the writing class is like, uh, yeah, I mean, we basically, we, we, we have a curriculum for that now. Um, and, uh, I think we're going to start picking that back up as like a formal thing of like, Hey, you can be in a cohort. Um, one of the, one of the things we found is this idea of like learning as a group.
is like very powerful. Um, you know, a lot of folks when we look at like, why do they stop doing kind of self paced stuff? It's because well, I'm not, it's not sort of like the pure accountability. So we found there's a big social element of this of like, okay, cool. Like we're all going to lean into great. We know we've got to take this, like, you know, we gotta, you know, we're gonna take this right in class.
We'll level up on this. It's like, okay, great. Well, why don't you just get a cohort of, I don't know, it could be 20, 50, 200 people. And we're all going to do it together. Uh, there's like real camaraderie around that.
Logan: On the writing side, like, is there anything that.
Logan: In, in, in studying this and turning it into an [01:22:00] actual practice, is there anything that's been particularly interesting or anything that, that you would impart on people listening about like. business writing and the style and not the five paragraph essay that you alluded to, but like what, what's been, what's been perhaps interesting in turning this into an actual program.
Andrew: I think the two biggest ones are one is getting people in the, you know, just getting yourself in the habit of writing down your thinking more often. Um, you know, so we have a norm internally. Uh, you know, that it's like nobody wants to go sit through like slides for 45 minutes. And by the way, it puts a lot of, uh, it puts a lot of pressure on the person presenting because like, God, I hope you practice last night because if you say the wrong thing, I mean, it's not written down anywhere.
This is your one shot. I think that's kind of a silly way to go about like sharing ideas, especially if they're important. Um, so one of the big learnings there was, uh, just a lot of folks are like, okay, cool. I don't actually know where to, if I have a thought or an idea, I You know, a lot of companies have gotten into it's like, Oh, okay, well, I'm so used to like the slide slides being the thing.
It's like, Well, no, no, actually, why don't you write down your thinking first? And actually, even stream of consciousness is fine. Um, a lot of people get very nervous. You know, it's kind of a blank canvas problem of like, well, if I start, you know, if I write this down, it may not be very good. It's like, well, actually, it doesn't matter.
Like, just, you know, it's a little bit, just get draft one done. And then the editing process is actually a lot easier, right? I mean, it's what makes it better, but your ideas will be out there. And then, yeah, we can, you can get feedback. We can hear them. So one is just building this into your ideation or you're like, hey, I have this thought.
How do I do that? Part of realizing too, it's that, you know, whether you're writing it in a, you know, Word doc or Google doc or an email, it's not really different. Like you just want to spend some time and, uh, so email writing is not all that different. I mean, there's obviously like the quick back and forth, but it's not all that different.
Uh, we really try to avoid the long, you know, the big essays in like slack or teams. Uh, I think that's like sort of like counterproductive. You want them to have like a place like a linkable spot. So the practice helps and the other [01:24:00] is, is, you know, it's a little bit like that Hemingway app that's out there, but just right as if you were, you know, talking to, you know, a smart, you know, friend or family member.
Um, I think there's, there's a lot of like business jargon that floats in and folks feel like they need to include that for some reason. Uh, that was the other thing that I think was very helpful is just hang on, just explain it. You know, it's a little bit like that Reddit, you know, explain it, explain like I'm five, like just explain it like I'm a smart person that understands this and who will ask questions.
So that those two things, the practice plus like, let's strip out that honestly makes people lot more confidence, um, and better. And we've gotten feedback that it's like, man, nobody's ever really critiqued my writing sense, like high school, the way we're doing now. And for such a critical skill, it's like, okay, well, glad you're getting some
Logan: It's funny. One of, uh, I had a boss early in my career who was a very interesting personality, but he used to make me write all of his emails. And, uh, I, he would, he would make me print them out when he didn't like them and he would go back through and he would mark it up and strike it and the, it was so painful.
And I, I, uh, wouldn't recommend this as a learning skill, but it forced me to write very simply and just very direct and communicative. And I hated it and I, I don't want to work for him ever again, nor would I ever make anyone do that. Uh, that works with me. But it was actually a really good skill of how to send emails, how to speak clearly it.
I found that there's a very natural, uh, uh, desire for people to try to speak circuitously or like verbosely. Uh, and it's actually just not direct into the point into learning that skill. It was forced on me talking about eating vegetables. That was definitely eating vegetables. And then maybe some.
eating shit as well. It was, it was not something I would recommend, but it was a really good skill to learn of like how to spend, send emails, how to communicate their thoughts clearly and all that stuff.
Andrew: like just on, I mean, so first of all, I did, I was my, wasn't my boss, but my [01:26:00] mom did the same thing, you know, probably from, gosh, when was in third grade all the
Logan: your boss growing
Andrew: Yeah, very much so. Yeah. But no, I mean, it was, it was, you know, you'd print it out, you'd hand it to her and she would go through it and edit and uh, yeah, I think, look, a lot of this stuff we talk about the craft of whether it's writing or otherwise, um, you know, it's this.
this obsession about details and, you know, a little bit knowing like when, you know, when you're kind of over the line, we're sort of in mission returns. I think that's the magic. I think a lot more people, uh, you know, when you're talking about honing skills, you know, something, something we try to do is it can feel, I think folks sometimes say like, oh, it feels like a little bit like micromanaging, like you're too much, like kind of in the weeds.
I think for some of these core skills, Um, everybody should have that experience where somebody's really, you know, editing your work. Um, and if you're really open to it and they're, you know, they're, they're kind of, you know, they're a master crafts person, right? Um, that's like invaluable. So I think that's something, there's a lot of this stuff that, uh, I mean, recruiting is maybe another good example, like I obsess about the way that we interview people and cause I had this, uh, experience, you know, the first job, you know, I remember being kind of thrown in the deep end about like Hey, Andrew, uh, we need you to interview some folks.
So here, just, you know, figure out something to ask some questions, ask them in here, go get in a room and go do it. And being like, my gosh, this is such an interesting situation. Like I've gotten like, I don't have any experience on this. I've gotten like no tips. And we haven't roleplayed this or practiced.
It's just like, go do it. And you know, I think, you know, I probably was not very good for a while. I think a lot of these things, you get a lot better if you have somebody that's like really going to put you through your paces. And, uh, you know, look, I mean, obviously you look at things like sports. Um, you know, you watch all like the kind of behind the scenes things and you look at how much they just obsess over, you know, it might be the angle of a pitcher's arm or like, you know, the, you know, I don't know, the way a football lineman blocks and it's like, yeah, why don't you do the exact same thing, you know, but in business and yeah, whether it's writing or otherwise, I think it's just a super [01:28:00] important thing for folks to, you know, hopefully everybody has an experience, you know, maybe it's not as.
Uh, harrowing as yours, but like, you should, everybody should go through that, like, kind of like that, like that, uh, line of fire
Logan: Yeah. There's, there's, there's something about like, uh, I don't know what it is, but enjoyment of the process or the inputs into stuff that is true. And if all you focus on are the outputs, you're going to drive yourself crazy in pursuit of those. And it's all the little things that add up to the totality of, um, successes or, or not along the way.
Logan: Um, so it's an interesting, um, kind of meta meta point as well, I guess on the recruiting side, like what, what have you learned about interviewing or setting up the recruitment pipeline and, and all of the things around that, that might be interesting.
Andrew: first when we started hiring, uh, folks for to join Klaviyo. Um, I remember my batting average of the first six people I hired, I think like, I think one or two lasted three months. Um, you know, some decided it wasn't right for them. Some we doesn't, wasn't right for us. And I remember thinking, it's like, my gosh, this is such an expensive process.
and to be so wrong is like killer. It's just so inefficient. Um, so, uh, I talked to a, I met somebody who was in recruiting, um, And they, uh, they gave me this book. Um, I think the author's name was like, uh, smart. And it was, I think it was a book was called like top grade or something like that. Um, but it had this very like systematic process about how to do recruiting.
So set aside the operational side. One of the things I loved was there was a line that was like, look, purpose of recruiting is if you want, it starts by defining what is the, what is the job. And I think that's honestly like something a lot of folks don't always do. So a norm we have inside of Klaviyo is.
You know, before we even start recruiting for something, you say like, okay, let's spend some time and figure out, like, if we hire this person, what is their mission for the next, you know, year? Like, what does success look like? And let's be [01:30:00] really specific, because at the end of the day, like an organization, a company, it's like, it's a machine, right?
And there's everybody's playing their role. And it's like, okay, well, what do I need this, this person to do? And I think the more specific you can get, you can do a better job of vetting for that. So one is we try to be very specific about that and we'll kind of debate it out, right? Hey, is this responsibility, you know, this person's responsibility or is it belonged to that other team?
Um, so in really figuring that out and like, well, how much progress do we want to see made and how do we think we're going to go about that? Um, so defying that matters. And then, uh, the other thing I took away was if you can get it right 80, 90 percent of the time. You are like, like, that's the gold
Logan: Oh yeah.
Andrew: So it's not, you know, I've seen folks like, well, I, I need 100 percent of my hires to work out. And it's like, no, no, no, that's not right. Like it's, it's, it's, you know, it's, it's impossible, you know, this is, it's a little bit like baseball, right? You know, uh, 300 is a good batting average, right? Um, so you can bat 80, 90%.
That's a good number. Um, and then we just measure against that. Uh, so that I think helped a lot. Uh, the other thing that was a little bit
Logan: measure against people's success on the hiring
Andrew: yes, yes, yes, yes. And I think one of the things actually the 80, 90 percent really helps with is. Um, especially for folks that are, that it's, it's, I don't know, the first time being a hiring manager or maybe if it's like, you know, a lot of people take a lot of ownership over their, over their success and their perfectionist.
Um, side note is we have a saying inside of clear, it's like clear is a terrible place for straight A students because you're going to get some C's and D's and like it's how you figure that out. Um, so it is, but the 89 percent paradigm really helps. Because I think people are afraid to acknowledge like a mistake.
Um, hey, this didn't work out. If folks have been through it before, they're like, okay, gosh, this just isn't working. And you know, it's, it's, you know, it's, it, it could be, you know, it could be we didn't define the role. There's all sorts of failure modes, right? Like a person's not working out, like, um, team structure anyways is not right.
Um, but acknowledging that and then resolving that really matters. So it's hard because if you ask people to kind of self grade, like, are you, I mean, obviously, everybody will fit to the test, but just giving them [01:32:00] that buffer that it's like, okay, well, things shouldn't always work out. Like, so let's just be really honest with ourselves about what's going on.
The last thing on recruiting is, uh, I think the experience of like a candidate really matters. So, you know, we have this value inside of Klaviyo about being customer first, which is all about put yourself in the shoes of the customers. In this case, I mean, it's joining the company. And, you know, work backwards from there.
I had this experience that first company I joined, uh, amazing company. One of my big lessons from them was they hired me on campus. Um, and so, you know, the way college recruiting works is often, you know, it's, it's your senior year. A lot of the recruiting happens early in your senior year. So you've got, you know, you, you might go do the interviews in September, October, and then you have six, seven months before you actually graduate.
Um, and so you get this job and you're like really excited, you know, cause it's kind of like, you know, I mean a little bit like you're picking your next team and you're, you're proud of it. You want to show it off. And so they did a really smart thing. Which is like, great, we're gonna send you some swag.
And I was like, that's awesome. Like, cause whatever you send me, like, of course I'm gonna wear it around and like, I'll be your free billboard, right? Uh, but they made this huge mistake and they sent me this like, double XL jacket. I was like, well, this thing, I'm like swimming in this thing. I'm like, of course I can't wear this around and I'm like, Oh, what a painful mistake on a detail.
So when I think when it comes to like recruiting brands, I think that experience of like how you get somebody excited about a company, you know, all the way from when they first learn about it all the way to like when they start, uh, it really, really matters. So we have this norm, like I'm like, never miss us.
The like swag that we send people in the early days of Klaviyo, uh, I don't think we tell folks like you can't, you know, I mean, nearly as good, we couldn't afford recruiters or anything like that, like internal or external. Um, so you had to go do all your own work, you know, so I had this norm, like I need to get people excited about a lot of the cold emails I wrote if they weren't the customers were to candidates to get them excited.
And I had this process of how am I going to get somebody so fired up that they're going to be excited to put in [01:34:00] time to interview with us. And so I had this coffee shop that was around the corner and this whole playbook about like, all right, you know, um, I would invite you and say, Hey, I've seen some really great stuff, some work you've worked on.
I'd love to chat. I was willing to like, I will come to you. So you just tell me where you work and I'll meet you at the coffee shop outside. Or I've got this spot and I'll meet you there anytime you want. Right. I was very, very open. And folks, you know, almost always be like, no, no, I'll come to you. That's fine.
And this whole strategy around like, okay, well, how am I going to ask them about like, you know, what they worked on, learn a little bit, like, is there a good fit here, but also, you know, kind of correlate the work they've done to what we're doing and really customize that. So this idea, like this idea of getting people excited about what you're doing now is very one to one.
And now we have more ways we do that at scale. I think it's super underdeveloped, uh, you know, one of the questions I have when I interview somebody that's going to be building a team is tell me about how you, you know, one great folks you've hired, but also two, how do you go about sourcing folks? Uh, and if somebody tells me like, well, I wait for the hiring manager to show up and like they give me a slew of people and I pick right as if I'm at the deli counter, right?
It's like, that is not the right paradigm. Like you need to put in your own hustle, like team building is hard. Um, and so I think folks that have figured that out, like, I mean, it's not surprising. They tend to have the best track records of like, Oh yeah, they build great teams. They hire really well. Um, so that's another skill that we just like, you know, tell us, Hey, look, I know this is expensive, but yeah, I mean, if you're a manager, like 10, 20 percent of your time is just on, I mean, just getting out there in the community and hiring.
Logan: Does the person that makes the decision on hiring, are they also the person that assesses the success? There's, there's some principal agent problem about, you know, if, if I signed off on the person I might be more biased to. Try to give them longer rope so that it doesn't hurt my performance ratio of my hiring.
Successors that done independently.
Andrew: Yeah. It's a good question. We, uh, so. It's a little bit of both. Um, you know, I think that's, that's a hard problem to solve. It's a little bit where like the, the kind of, I don't know, the [01:36:00] entrepreneurial in this of like, I mean, you are asking people, that's why the eight or the 80 percent 90 percent kind of helps because you want to give people buffer for, Hey, Like, okay, like devil's advocate, let's say things aren't going great.
Like, what's the thing? Um, you know, I'm known for inside of Clavio is like I work on being a little more positive, but we're very quick to be like, okay, cool. That was awesome. Let's celebrate that. But then like, okay, what are the ways this could fail? Right. And we do that for projects for teams, we do it for people.
Like when I'm giving somebody a review. I was, you know, I look, it's like, look, I want you to be the best you can possibly be. And so therefore, these are some things that like, maybe they're hard to hear or not, but like if you, if you advance, you know, the ball on these dimensions, you're going to be even better, uh, than what you are today.
It's like, let's, let's, let's, here's how do we build, build a plan to go do that? That's a lot of the best feedback I've gotten has been from, you know, having folks, you know, advisors, um, you know, folks on our team give me that kind of like, yeah, Hey, you'd be better if. And so we try to do the same thing.
And like, I think if you do that, you can kind of figure out like, okay, like are some of these, you know, if somebody is like kind of not working, oftentimes I find when we have that conversation, sometimes you can unearth, it's like, oh man, there's like a big kind of, there's maybe there's a big issue there.
Okay. So what are we doing about it? Um, it helps us like dig into that a little bit.
Logan: Once people come in the doors, there's the, the element of, uh, onboarding and, and culture and, and all of that. And we talked about the different layers of individual contributor, manager, manager of managers and all of that. Um, culture's kind of the other side of. Uh, the the coin or the way that you espouse or codify some of the ethos that the company is going to be.
And what's what is culture is what is done when people aren't watching or whatever that phrase is. Um, what we we've talked about some of the cultural, uh, uh, premises that you guys operate on. Are there? Other things that you've you've learned about not not specific cultural elements of Klaviyo, but building a [01:38:00] culture and and having it manifest throughout an organization and some of the benefits that come along with it.
Andrew: One of my beliefs is culture is a product. And we say that because you can think of culture as like, you can either be intentional about it, like a product, uh, or you can just kind of let whatever happens, happens. So we say, no, culture is a product, we're going to be intentional about it. The other thing that happens with products is the goal is not to say, you know, you go talk to customers about what products they want, but then you have to, you know, you have to build the product that actually solves their problems, not necessarily what they ask you to build.
Um, I think it's very important. I think that the two biggest, uh, the two biggest failure modes I see with culture or like, you know, inefficiencies is one, not being intentional about it. You don't have a roadmap the same way you would with a product. Uh, and then two is overfitting for, hey, well, this is what I'm hearing, you know, from surveys we run, et cetera, and then just literally taking verbatim, like, what should we do?
It's like, oh, well, we should have program X. And it's like, hey, let's go do program X. Right? Um, instead you say like, no, this is what we stand for, uh, and therefore, like, we're going to then go build a culture or programs that are around it. So culture is a product. Then as part of that, uh, two other things that maybe we're a little opinionated on.
Uh, the first is when you think about like, you know, people, that's so much on values and all this kind of stuff. Two things I think people often get wrong that one, you know, values that a company has and we think of those as like the way you want people to act, work, behave. Um, those are in support of a mission.
I've seen a lot of companies where it's like, well, we kind of made them and you're like, okay, but why'd you pick that? And the answer that I'm always looking for is. Well, because that will advance our like mission. This is our purpose is what we're trying to solve for. And that's the reason it matters. A lot of people like lose that linkage.
So I think that's very important. Um, and then there's like, there's some evergreen things that everybody should do. For instance, every company should care about their customers. You know, it's a little spinal tappy. It's like, I think the difference there is there's companies that sort of say it, but they sort of turn the dial.
It's like a two. And there's companies that like turn to an 11, right? So this, we talked about this, [01:40:00] you know, uh, Hey, engineers are responsible for doing support. Like, there are not many companies that, that, where that's a norm. And that's like one of the examples where we're trying to like, okay, how are we going to turn this up to 11 and really be extreme about it?
So I think you need to, you need to figure out like, one, it's a product, figure out which, what you stand for and then like what those things are, why they tie back to a greater purpose and then turn it up to 11 and 11 is we just think of like, what are the rituals that we have? Right. So, you know, the customer first one is like, yes, spending time with customers.
Right. You just got to show up. Um, there's a lot of other ones about, um, you know, uh, you know, we, we highlight, we talk about doing remarkable work, high quality work, people don't want, people just want to rave about it. Every week we highlight a project, I mean, one of the great parts about being 2000 people is there's all this serendipity of great work happening.
You just have no idea it's there. So we have this like every week we highlight a project somebody's done where you're just like, Oh my God, that like is literally something you'd want to go home and like tell your friends and family about. And that takes up actually a bunch of work to source because it's like hard to like find those stories.
Um, and get people to tell them, uh, but building those in is like that you have rituals that reinforce those. I would think like when you talk about folks and you say, Oh, X, Y, Z has such a great culture and you say, well, explain it. They'll always explain through examples. Well, here's an example of when, or here's a ritual that we have.
So I think if you do those things, like that's, that's kind of the path to really defining it and treating it like a product.
Logan: building a company in Boston. Obviously, Boston's had a ton of successes over the years. Um, but, uh, there's this element of, of companies outside of Silicon Valley and New York and Boston, I think suffer elements of this maybe less so now than they have in the past of just, uh, and it's certainly true of companies that I've seen in Atlanta or, uh, Raleigh or, uh, Denver or whatever it is where, uh, When you're at the scale you guys are at now, it's when people leave and they talk to their coworkers or [01:42:00] their friends and families and all that, like you guys are in rarefied air in general, right?
But certainly in the Boston ecosystem now there's Wayfair and there's HubSpot and there's, you know, other PTC or whatever it is, but like definitely rarefied air. But I can tell that your ambition of where you want to take this Is nowhere near what the business is today when you're in the Bay Area, there's there's or even Seattle, there's trillions of dollars of companies and equity value.
And you know, if you're not at open AI right now, probably in the Bay Area, everyone else is sort of like, Oh, gosh, well, we're not the cool kids or whatever. But I assume a lot of your employees are the cool kids for. You know, they're, they're, they're friends and all that. How do you maintain, how do you use Boston to your advantage or that culture, but also maintain the ambition of, no, we want to be a hundred billion dollar business.
We want to be whatever, hundreds of billions of dollars, uh, for your employee base and have that manifest all the way through.
Andrew: Yeah. Um, well, so I mentioned this, you know, having this mentality of, you know, the progress bar is always at 1%. And I think that you got to think about, well, what would it mean to be a hundred? What was a hundred X where you are
Logan: I know you like abstractions, uh, like higher and higher
Andrew: So you got to eat. So no, and it's, it's interesting. I think one of the things that happens to a lot of companies, uh, after, you know, there's always this kind of like existential dread when a startup company starts of like, Oh my God, we're not going to make it.
Um, and you know, I'd say there's a little bit of this, like, you know, thing that happens to for some companies, it's like, Oh, this, you know, it's a fundraising round or it's going public, but it's, it's these kind of artificial moments. Um, and I think what those often, what those can expose is, well, is there this greater ambition?
Like, do you have this point of view of like, Hey, in five years, 10 years, we could be, we could do this? Um, you know, for us, it's about being this like scalable brain, the scalable interface, you know, way that any business organization can interact with any consumer, right? Um, you know, I think about all the, the dozens of organizations, you know, whether they're retail brands or any kind of company that I interface [01:44:00] with.
And I'm like, yeah, I want every single one of those interactions to be awesome. Like I want to love it. And I want those companies to feel like they've got like the tech where it's very cheap and easy for them to like go deliver that. Um, so you gotta have, I think you gotta have that ambition. I think that just, obviously folks just gotta step back and you know, you just gotta think it through, right?
A little bit like our five year plan thing. Um, the other part is, it's interesting, like we've, uh, I think there's a lot of value to like brand PR, um, and especially when it comes to like, you know, employees, uh, we have taken a tact of humility matters. And so in some sense, it's if you believe in what you're trying to build towards and that's the exciting thing and you're kind of like, you know, shoulder to shoulder with other people to care about it, you're, you know, getting in there with like the people that are actually users, customers, partners.
The rest is kind of just noise and it's fun noise, don't get me wrong, but it's kind of just noise. Um, but yeah. You know, for a lot of years, Cleo, we didn't, there wasn't, well, we didn't have a PR function, but we just didn't do much press or really anything. And it's interesting because I always had this kind of like double point of view about it where I was like, ah, kind of lamented because you're like, ah, it's not like there's just everybody's like lining up to, you know, interview and all this kind of stuff.
But at the same time, we built an awesome team, um, and a team that was very committed, you know, to each other and towards what we wanted to build. Yeah. And I, I think when companies get more successful, there's actually this real risk that all of a sudden it's like, well, there's folks that want to join just because it's the cool thing.
Um, you know, I had friends that, uh, you know, went to the Bay Area and, you know, when I talked to them about like, oh, so what's going on? Like oftentimes they talk about, it's like, well, here's where the cool thing is. And, you know, I'm trying to decide if I should join it or leave it, but it was very like, you know, every 18 months, every 24 months, there was a new
Logan: Yeah, ephemeral.
Andrew: Yeah. And I'm like, that's exhausting. Don't you just want to like build towards something? I mean, like, look, if something's not working, sure. Like go, you know, of course, like you can experiment with different things, but it feels [01:46:00] exhausting to play this kind of like social game of like, what's the most exciting hyped thing right now?
And, you know, again, like a lot of it's, I mean, a lot of these great companies, I mean, they're, they're real, like opening eyes are real. That's real tech. Um, but it's like, but the reason to join is because, you know, let's say, I don't know if the mission there is around like, you know, uh, you know, LLMs and AGI, then great join for that.
But, you know, do the mental thing of like, well, what if that whole thing, you know, I don't know, what if, what if things don't go as well as planned, right? What if, I don't know, GPT five takes a lot longer to build than we think, like, would you still do it? And if your answer is no, then like, well, gosh, maybe that's not the right place to do it.
I, I think actually it's gotten harder for us to recruit because it's a constant check you have to run of, you don't want to be the, I mean, sometimes I'm like, we don't want to be the cool kids. We want to be like, we're here to do our thing. And it's sort of, I often try to, you know, how do I, how can I sell against that?
Ah, we're actually not going to be a cool company in five years. And that's like, would you still want to be here? Cause if you do, that's probably a good fit. If you don't, then like, I'll save you some time. Right. Go try something
Logan: Interesting.
Logan: I was curious about, you've talked about productivity as software versus output as software, and I assume moving more to artificial intelligence and that we're going to move less on the pure productivity and more on the outputs. But I guess, was that actually a founding tenet of Klaviyo in the early days?
Andrew: Yeah. I mean like I had a, um, you know, I mentioned that, so I worked, I think, you know, if you're, if you want to be a founder entrepreneur, right. Um, I think there's a lot of mythology around, Oh, I have this idea. And the media does not help with this. Is like mythology around like, I had an idea, I like dropped everything, quit my job, dropped out of school, started working on it, and it just poof, worked.
Logan: struck
Andrew: Yeah. The reality is if you talk to almost any successful entrepreneur or founder, they'll be like, oh my god, it wasn't the first one that worked. It's like, are you kidding me? I've been experimenting with ideas for years. Right. Uh, so for me, I mentioned it's like, [01:48:00] okay, so I did a little bit of that experimentation inside of other companies, but you know, then nights and weekends I was working on other stuff.
And I have this, um, so I'm a pretty big runner and, uh, I, you know, one of my pet peeves is it's like, there's no great search engine for finding five Ks and marathons and this kind of stuff. It feels like there should be, but there's all these like very old websites, like, uh, early two thousands era. So it's very hard to filter.
So, Um, anyway, so I, I have this idea of like, Oh, I'm going to build a search engine for that. And then I'm going to, you know, maybe I can get some of these races to advertise and like that could be a business model. And so while I was playing with this, I, uh, you know, I would, I made a list of all these like little local five Ks or, you know, maybe it's like, we're, you know, we're here in New York, it's like New York marathon, the Boston marathon, and like to talk to the big, big events too.
And I was doing all these calls and anybody that's, you know, look at anybody that's been on a sales, sales team, BDR team has done the same thing, right, and do all these discovery calls and I'm just going through one by one by one. I was like, this is insanity. I like, I more or less have my playbook. Like, I know the kind of things, questions I want to get.
Like, this ought to be the kind of thing that like, I want to provide a good experience to somebody, but like, I do not need to be doing this myself. I can't send them a form because it's way too impersonal. I'm like, why can't I? Like, why isn't there a way for me to just take a part of myself, clone it, and be like, look, I want you to talk to all these people, like, maximize for whatever we're trying to solve for.
In this case, maybe it's information gathering or whatever. And uh, but I want it to feel very human. Like I think we're not far off from. You know, when you, you know, somebody could call, call or you could do a zoom and it's like, it's not actually me there or you. It's like our surrogate, it's our like avatar, but it is us in terms of how we think and how we want to act.
Um, and this is certainly works for a business where a business effectively like, well, there's no one person, right? A business, you know, they write these training manuals on, this is our, this is, this is, these are brand values. This is how we behave. So, but I think that can get, get encoded in software. And I think when you do that, what's, what's interesting is you're like, well, how would you know if it's any good, if it's being successful?
And it's like, well, if you can measure some output metric, I mean, all of machine learning is based on this idea of like, [01:50:00] there's some reward function output metric that we're like trying to maximize for, right? We're fitting our regression to or whatever. Um, I think if you have that, I think we're not far off from having software that You can say, look, I don't want to treat this as like user goes in and just, you know, put pecks at the buttons based on what they think is the best way to maximize for whatever they're trying to solve for.
Just have the user declare what they're trying to solve for. And then like the software should just run itself. And uh, I have been a very big believer of that for like the last 15 years. Um, you know, we definitely have a lot of the data to do that. The hard part actually is a lot of applications is finding the, like being able to measure that, you know, output function.
You know, we'd go back to writing, like, good God, like, how are you going to measure like, is this a good email or not? It's like, well, I don't know. I mean, like somebody needs to score it, label it. Um, Now it happens like for the domain that we're in right now, um, marketing, it's much more measurable. You can measure conversions and that's much easier.
Um, but I do think that's a paradigm that's coming to software. So we're very interested in this. Like how do you do attribution? How do you do this reward function discovery? How do you then build the systems to take all of the like dependent variables? And correlate those and how can you make the machine, you know, make the software learn on the fly.
Um, I think that's very interesting. The uh, what's, you know, the, we'll see if the LLMs really help with this. Maybe they do, right. If they can get to reasoning. Um, they've been especially helpful in our domain on like the content creation side. Like just, you know, that, that's cause that's a real part of marketing.
But yeah. Anyways, I do think that's like, you know, the Steve Jobs quote of like, Hey, the, you know, the, uh, by, uh, the, you know, the computer is like a bicycle for the brain, right? It's a great one. But it's really, it's like, it's, it's, yeah, it's like productivity tool. I think we're close to because we can measure these reward functions.
We now have the algorithms to do this, that like we can discover some of the algorithms that we ourselves are using. And I think it's super interesting that we talk about learning, right? I mean, so much for us is guess and check. I think it's very interesting. It's like, my gosh, but what if we could take the algorithms we're using today put that in statistical software, have it, you know, then go kind of train up and learn, like maybe it could [01:52:00] learn faster than we could, maybe it could teach us back, you know, how to, you know, how to improve things.
So I could have a business and say, This is the way I think I should treat my customers. It's fascinating. Like when I talked to a lot of brands, you're like, yeah, I think that's the right idea, but I don't know. I'm a little bit making it up as I go. Uh, one of them said, it's like, okay, cool. You tell us the ethos, what you want to stand for.
And then we'll really, you know, we'll harden it. We'll optimize it. Anyways, I think we're, we're close to a lot of software doing that. And we're, one of the things we're working on is the patterns by which, um, you know, software of the future will be built to accomplish that.
Logan: Super cool.
Logan: Well, Andrew, thanks for doing this.
Andrew: Cool. Thanks for having me.
Logan: Thank you for listening to this episode of the Logan Bartlett show with CEO and co founder of Klaviyo, Andrew Bialecki. If you enjoyed this discussion, we'd really appreciate it if you shared with anyone else that you think might find it interesting as well as subscribed on whatever podcast platform you're listening on.
We'll see you here next week with another great guest.