Y Combinator Startup Podcast - #58 - It's Surprising How Much Small Teams Can Get Done - Sam Chaudhary of ClassDojo

Episode Date: January 24, 2018

Sam Chaudhary is the cofounder and CEO of ClassDojo. They've raised $30M and have 30 employees.Karen Lien is an Edtech Principal here at YC.The YC podcast is hosted by Craig Cannon.

Hey, how's it going? This is Craig Cannon, and you're listening to Y Combinators podcast. Today's episode is with Sam Chowdry and Karen Lean. Sam's the co-founder and CEO of Class Dojo, and Karen is an ed tech principal here at YC. So Class Dojo is a communication app for the classroom. They connect teachers, parents, and students who use it to share photos, videos, and messages throughout the school day. And in this episode, we mentioned Imagine K-12, so IK-K-12 is an ed-tech accelerator, and they now make up YC's EdTech Vertical. All right, here we go. Well, I don't want to miss this story.

You've got a sly grin. This is a stringy best story. So, this is not the stringy best story. So little known fact, one of your first investors was Paul Graham of Wycopinator freedom. Can you tell us about that meeting? What convinced PG to write you a check? Yeah, it was hilarious.

The whole thing was hilarious, yeah. So it was actually, it started at Demo Day. And I... Imagine K-12, Demo. Imagine. PG was a guest. Invited guests. It was lucky to be invited. Yeah, right.

Not running the show. Yeah. So we'd presented last. And we'd been told the whole time that, like, you know, the goal was to meet lots of people and to get them to come by and, like, talk to you, right? And so a good way to do that is to have like good metrics and we were lucky we had this good growth curve. But another way to do that is is donuts. And so I had a box of donuts and I said at the end that like, hey, we're over there. There's a box donuts on

table. Like, I should come by. And so, and so the presentation's finished and we were kind of milling around. We were at our little stand thing, hoping that people would pop by. And who walks over, but PG. And he goes to Liam. He's like, hey, it was one of the metrics we showed. I think it was an engagement metric. He's like, I really like that,

like the look of that, like. Do you remember what that was? I don't remember the exact one. I'm sure I've got it in the, I think it was something about every three seconds. Oh, yeah, people giving feedback or something. Rewarding feedback. Yeah, maybe.

You remember it better than I do. Yeah, I don't actually remember it. I've got it in the pitch deck somewhere. I'm sure you guys have it too. Yeah. And so he's like, like that metric can you show me like the what the growth curve in that looks like and um and liam like i can't even imagine what was happening for Liam because he and i we like read all of pg's essays for ages right

this is like like a legendary person turning up and just like having a normal conversation with you and and and so poor Liam was just like kind of like a holy shit you know he's like oh my god like my my my idol is here or whatever um and Liam's like well i i i don't have it to hand but i can pull it or whatever and pg's like oh yeah you could just open tunnel and do it in there, right? And like, he's like, starts coaching him through it. And Liam's like, like, he's an amazing technology. He's very, like, he's an amazing technology.

He's very competent. And he's like, oh, yeah, yeah, I guess I'll just do that. And so, but it's going to take a while. And the donuts kind of, kind of bit us in the ass where PG's like, that's okay. I'll just stand here and have a donut. And so you've got this like high pressure like data extraction happening for Liam. And in the end, Liam's like, look, like, better than this, I'll just email it to you, like straight after this.

And so Paul gives us his email address, and he's like, that sounds good and walks away. And so we go home, like, breathlessly excited. I don't remember, like, the rest of what happened at Demer Day. And we craft this email to PG or Liam does. He's like, hey, here's the metric you want to look at the growth code or whatever. And we send it off. And, like, nothing, like crickets. And we're like, oh, I guess he wasn't like that interested.

Like, he was just polite, right? But being relentless with this stuff, two days later, we email them again. And the numbers are small so they're growing. We're like, look, it's growing like three X. And again, like no answer. And we keep doing this for like, I think the best part of like a week, like every day or two. We're just emailing like another chart, another chart. And then from Liam's email.

And then I get an email from Paul like a week later with one line network. He says it is customary to respond to offers of funding. And we're like, what does this mean? Right? Indeed, it is, yes. Yeah. And we're like, yes, agree. Do you want something else with that?

And then we check out, I think Liam has some kind of super juiced up kind of Gmail inbox or whatever. And it's like, it's like in a spam folder somewhere is like like two days in. PG is like, okay, I'm in. And like, whatever, I'll write you guys a check and just come around and pick it up. And we're like, so we basically spammed PG for the best part of a week after already had a number of funding. And then, yeah, then Liam biked to his house to pick up the check, which was kind of cool. Yeah. I think he was having a nap in the shed.

That's great. Tell us about who uses Class Dojo and what problem they're trying to solve when they adopt Class Dojo. Yeah, I think it's important to understand like why Class. even exists, and then that gets like the problems that we're actually helping people solve. So the reason we started the company really is because I kind of thought that most kids don't get the education that's going to make them be happy and successful. That's bleak. That's bleak, yeah.

But what's happy, the good thing about it is that I think like the answer is already there. Like the answer isn't like some magic technology or magic policy. It's actually like education is really a product that's made by people in classroom. and in homes. And so our job as a company then becomes to help people in classrooms and homes create a better education experience for kids. And so, like, that's really, like, what class teachers have been trying to do for a long, long time, like, from the start of the company.

And so the exact set of problems that we've tried to solve changes quite a lot, right, depending on the classroom, depending on where kids are, what kinds of things teachers want to do. But I'd say, like, probably the two main problems we solved today. day. One is that there's this, there's this weird, like, golf between school and home. So, like, kids go off to school for, like, eight hours a day, and a bunch of stuff happens in a classroom, and then they go home, and a bunch of stuff happens at home, and, like, the kind of the two halves of the day don't really, like, connect and don't really talk to each other. But that's

really, like, a kid's life growing up. It's, like, usually, like, one of those two settings. And it's a bit odd, because, like, in every other part of your life, you're connected to the things you care most about, like Instagram or whatever, but it's very odd for parents that they're not more connected to their kids for, like, most of the day. And for teachers who actually really care about these kids that they don't really get a view of what's happening at home. And so we really, like, that's one big problem we try and help with is, like, bridging that golf and, like, creating more of a connection between school and home. And then the second problem, there's a newer one, but it's really this, it's kind of like teachers have struggled for a long

time to try new ideas in their classroom. So like the pace at which new ideas get introduced to classrooms is actually quite slow. And it takes quite long time for a teacher to try a new practice. They might have heard of like a personalized learning or like something like a growth mindset, like a new idea. And it's quite tough. And so we're trying to like help teachers try more ideas in their classrooms and make it easier and faster for teachers to bring new ideas into their classrooms. How does your product make that easier for them? Oh, yeah.

So the first part, we basically just helped teachers share more of what's happening at school every day. Sure. So teachers could share pictures and videos through the school day home. So this new bit was from like a couple of years ago where, and it kind of, it wasn't really like a plan. It kind of emerged from something one of the teams was doing. So we spend a lot of time in classrooms and always have. and we kept hearing this thing about social and emotional learning. Like there are lots and lots of teachers who really want to teach their kids things beyond

like reading and writing in math, but like softer skills. And it was really tough for them because it's like quite a fuzzy area. They're like, what do we do next? Like I want to teach my kids about curiosity or creativity or empathy. These are like important things for people to learn, but we don't really know how to teach them. And so a few years ago, we went down the road to Stanford, and there's a professor there, a Carol Dweck, and she's got this, you know, right? I don't know if everyone listening knows, but you know. And she's got this idea called the Growth Mindset.

It's like a really famous TED Talk, and she's got a book about it. And it's basically the idea that, like, you're not fixed in your abilities, and that through practice and persistence, you can get better at stuff. But it's amazing how many kids grow up without that idea really in mind. They're kind of told you're a math person or you're not a creative person. These are really quite limiting mindsets. And so we wanted to see if we could help teachers, like undo that. Like just start with that one idea, see if we could help more teachers share this idea for growth mindset. Because they already talked about wanting it.

Like we heard it from them. And so we went to Carol Dwack and her team and we were like, look, if we could get this idea to lots and lots of schools, what would we do? And together came with this idea of like making short stories, like animated shorts, three to five minute little thought starters and conversation starters for classrooms. And so we produced, there's the most like heck job I think we've ever done. It was like eight weeks from like the first conversation to getting this series done. We did it all in house. It was like three people in like a tiny room.

but we made this series of like five animated shorts about growth mindset and they they weren't really like you've got the Khan Academy kind of lectures they weren't read lectures they're just like stories and at the end of the story there was a question so it didn't really give you the answer it's just like the story like teeing up some interesting thoughts for you to discuss and so we made this series and we put it at we distributed it on class dojo so every teacher wrote a class dojo account it popped up in the app one day for them it's like hey you can teach your kids about growth mindset and it was crazy because we'd never done anything like it before, and we didn't know how it'd be used, but we ended up

with something like, it's something, I think those videos have reached about like 15 million kids now in classrooms, which is like nuts. And so that was the first time we ever did anything about this, like bringing new ideas to your classroom, and then we did a follow-up with Harvard on empathy, with Yale on mindfulness, and there's more kind. So it's literally the teacher says, we want to be able to do this, and you say, we'll create it for you. Or we'll find experts.

We're like, we're not the experts on growth minds, but there are experts on it. And if we can... It's not just facilitating things that happen in the classroom. It's actually providing. It's actually helping people move classrooms forward. If you really believe the thing that I said that most kids don't get the education that they should get for now and the future. And do you believe that teachers want to do the right things for kids? Then I think it's important to help teachers not just do what's really been done more efficiently, but actually help them do new stuff.

Yeah. And how are you measuring the success of teaching a growth mindset? Yeah, it's a really good question. So, like, there's kind of two, there's like an input and an output measure, right? Yeah. And the input measure for us was really just like, do people use this? Okay. And so use meaning like use the app?

You watch the video? Yeah, do people watch the video? Do they do the activities in the classroom? Do they share at home with parents? Yeah. And so that's quite a good like effort measure, I think. Okay. The real truth with the growth mindset stuff is that like the output measures, I think, are like,

from what we can tell from research, they like take a bit longer. They're mostly like psychology research and it takes a bit longer to really see if people's mindsets have changed. So I think like there's this question that comes up a lot of like should you do stuff that people use? Should you do stuff that's like like high efficacy? I think it's actually like a false tradeoff. But I think you have to be clear about what today we know how to measure and we don't know how to measure. And the thing that like we can measure is like are people voluntarily adopting and using this? Like is their demand for this?

And the thing that's going to take a little time to figure out is like, hey, are people, is the whole world now having more of a growth mindset. But what we do know is that, like, for 50 million kids, they've been exposed to the idea for growth mindset, and they're using the words in the classroom, and they're using things like, you know, it's not that I can't do this, I can't do this yet. And, like, that's kind of a cool, like, stunning point, I think.

I think we're far from done with it. Yeah. Great. So from the beginning of the company, you've been very effective at talking to your users in such a way that you really understand their problems, I think. And you're describing now, you know, as the company grows, you're gathering this from your users and giving them the things they really want for

their classrooms. Can you talk a little bit about how in that early stage before you knew what your product or even maybe your audience was, how you figured out, how you developed, like, what are strategies for developing that deep understanding of your audience? What are some specific things you did early on? and that you continue to do to keep that connection so you can be building the right things, solving the right problems. Yeah, that's a good one.

It's a really important, like, theme in the company. We talk a lot about empathy a lot. I think, like, so before we even came up with, like, the idea for the first product we ever built, actually, like, I think something people might not know is that, like, class data didn't actually start, I mean, you remember. We didn't actually start as, like, a company. It came really from, like, I'd worked in education. for most of my life, like in classrooms, around classrooms.

Liam had been, like, co-founder had been doing a PhD in computer science, focused on technology in classrooms. And so it really came from more like a passion for education-y stuff. And when we turned up in California to do Imagine K-12, like we basically didn't know anyone in America at all. Like we'd never lived in America. We never worked here. Like we'd met like a guy called Jeff Ralston on like a video called once, right?

And like Tim and Alan were on there as well. and so it kind of enforced like a naivety in a way like it kind of removed any sense of like hey we know what we're doing and we'll just we've got the idea and we'll go and do an idea so it kind of enforced that like get out and like try and understand what people who are actually doing this need and so the first like four I think it's like six weeks yeah we heard for six or seven weeks before we launched the first version of class dojo seven weeks And in those seven weeks, it was mostly me, but Liam was involved too. I did basically anything to get in front of teachers to get to classrooms and talk to people.

And so on the easier side, it was some schools published teachers' email addresses just on the website. And so I'd like just email people. And it was kind of a weird email, right? So it wasn't like a pitch for anything. It was more, hey, like we're the, like I told them a bit about us. and I'm like, we'd love to help in any way that we can with whatever you're doing in your classrooms. And like, so what's the worst problem that you face every day? Like, just please let us know.

It's like asking for like a one-line response. Just let us know what the worst problem you face every day is. And so we did that. I think I emailed a few thousand people and it was like scraping email addresses. It was like, I've had a friend at Teach for America. I got on some of the email list there. We had friends who were teachers in the UK and Australia. We kind of emailed just as many people as we could.

But the other thing, which was like more, probably like if you'd step up the difficulty a bit, was we'd go to local schools. And so, like, we were in Palo Alto. There was a gun high school nearby. And I found a few teachers there who were willing to talk to me. And I went and taught summer school there, in fact, for like a couple of days. Just like, they promised to like, like, hey, if you can, you've taught if you can come and teach like summer school for an afternoon. And sure, we'll chat with you after summer school. I'm like, cool, I'll do that.

we went to like teacher meetups. There are these things called the ed camps. There's actually one ed camp, which was really like life-changing first, which I'll tell you about a minute. But like we went to these teacher meetups and we're just like as much as you can to soak in the context of your users.

Like that's what we did. That's what I did in the early days for a lot of it. We had a particularly horrific one once where we had to go to Los Altos school district. And we were like pretty stingy Brits. And so we didn't have like, don't want to get a cat. Like Uber wasn't around then.

We didn't have a, get a cabs. We, like, biked for miles in, like, the steaming South Bay Sun to get there and turned up just dripping with sweat. There's audience of teachers. It was really, really, really, really, yeah, really lovely. Great intro. But you kind of, like, you kind of just do whatever it takes, right? Like, when you're two guys in a room and you need to get to understanding, like, how you can actually help people, you do whatever it takes.

So I spoke to, I think, three or four hundred people on Skype calls face-to-face in those seven weeks. emailed a few thousand more. And as the company's kind of grown, that's just become like an institution in the company. Like it's, I think, you know, we're only 30 people, but we may be unusual in that we have, like, research as a function in the company. Like, there are researchers in the company who spend all of their time talking to teachers. We have, like, lots of ways of bringing teacher context in, and kids and parents now as well, but bringing that into the office. Like, it's up, like, the story's up on monitors. Like, we have. have community groups,

so we solicit feedback all the time. So I think that that's just like it's an important part of how we build class edger. Is that a vestige of you working in consulting? Or do you think it just like fits the company? Because I saw that you worked at McKinsey. And this is a very common question that we get like, hey, I'm working in consulting, thinking about doing a startup.

What do I do? And it's like this consultant versus entrepreneur, like get your hands dirty mindset, right? Yeah. Yeah, mine is a bit weird because, like, consulting was kind of the anomaly for me. Okay. So I was in my teens, and Karen, you know this, but like in my teens, the school I went to actually insisted that you teach as well as learn. So I ended up teaching for like 20 hours a week for six years from age 12 to 8 to our peers.

To our peers, yeah. Okay. And many of us did. And it was crazy. But it's also one of the most effective things you can do. It's like, you know, the high cost, low cost, high quality interventions is peer tutoring. And so, like, we had this in our school. We had, like, I would teach regularly, teach classes of 20 to 30 kids for, like, around 20 hours a week.

What was your subject? That's all of them, right? Like, you go across. Yeah, yeah. I mean, you'd go across, like, physics, chemistry, economics, maths. It's a really, like, one, like, really fascinating kind of school. But so I taught for, like, a long time. And I got to college and I was an economist, like a math person.

I thought I would carry on with a PhD in that. And I ended up. going to teach straight after college instead. And so like all the conventional consulting banking didn't really appeal that that much. But then McKinsey, you're very good at the CRM system kind of follow up. And they're like,

they track where you go. And I was like, oh, I'm going to teach. And they're like, we've got this education team here. You should come and join the, you know,

like do some education work. Really? And they were like kind of advising governments on how to set up education systems. And so I went there and I spent a bunch of my time there in education. And ultimately the reason you just said is the reason I left that like, I think there's only so long, like if you're used to like,

getting your hands dirty, there's only so long you can, like, advise people and not actually do the work. Yeah. And so I think that was more, like, the way that I'd always been consulting was, like, the kind of the left turn bit. But yeah. Okay. It was good, but it wasn't for me forever.

So you talk to these hundreds of teachers and released this first product. Tell us about the early days of growth, what you think drove that and how that's shifted over time. Have there been transition points where class dodo went from kind of this growth engine to this other one? Have you anticipated or managed those transitions and what do you do now? Yeah, no, it's... But tell us about that early growth. So we had our first like seven weeks or six, seven weeks or so of talking to teachers, like building like little prototypes and things. And then week eight, I remember this because it was a Sunday morning and we'd put the first version.

of class digit, like, live for anyone to sign up for. And it was quite early on Sunday. It was something like, I think it was like six or seven in the morning. And there was this ed camp in Oakland. So an ed camp is really just like a group of teachers coming together at the weekend to talk to each other about, like, new things they're trying out. So it's kind of like teacher professional development camp. But they're doing it at Skyline High School in Oakland.

And we were in Palo Alto at the time. and I work over at 6-7 and I was like, oh man, like I'm exhausted. Like we'd been sprinting all week to get this product out and I was like, I should probably just blow this off and like, be nice to have a lion. And I don't know what it was, but I posted, I think it was on Facebook and I posted like, I also didn't have a way to get there.

I was like, I can't bike that far. Like the cab's going to be really expensive there and back. And so I posted on, I think it was Facebook. And I was like, does anyone going towards Oakland today, like on the off chance anyone's going? I'd love a ride. And then it turns out another person from our Imagine K-12 class, Chris Streeter, in fact, was.

And he was like, yeah, I'll pick you up and go. And I was like, oh, I guess I'm going to Edcamp now. Yeah. So it's a good lesson to, you know, try to be the hardest working person in the room, right? But like Chris put me up, went to the ed camp. And based just starting, there was like 80 or 90 teachers there. And we were just chatting. They're like, oh, like, you're not a teacher.

What are you up to? And I told them about this thing that we've been working on about class dojo. And they're like, oh, that sounds really. really fascinating. And then like two or three of them were like, oh, I might, I might like check it out and like tell some friends about it as well. And so we kind of had these two things. Are the people we'd already spoken with who you can imagine like the most magical experience you could get is someone turns up and says, hey, we just want to solve a problem for you. And then goes and build something that tries to solve the problem. Like that's like a good

experience for you. That's like a very white glove kind of service. And so a bunch of people got that. That was the 400 teachers or so that I'd spoken with. And then this. new set were like, oh, that might work for me as well. And then they started using it and they like Twitter was actually big for us in the early days. They started telling like their teacher followers and friends about it. A few of them turned out to be quite like influential, like thought leader type teachers. And so by the end of that first week, we had that something like 80 teachers using it every day. And we're like, well, this is amazing. 80 was like just an unimaginable number. And this is in the summertime when...

This was like right at the end of the summer. It was like I think we got to the last week of July or first week of August. Okay. So schools are not really in session. Well, you've got summer schools. This is one of the secrets was like teachers are already preparing for like what's coming at the end of August. It's already trying things out.

They're already like setting up classes and all that kind of stuff. So this was like kind of crazy. We're like, wow, there's people like actually picking this thing up. And then what happened was from then through Demo Day, which I think was like, Was it end of August? Early September. Early September. Yeah.

So it was roughly a month, like four or five weeks or so. And it had gone from like 80 teachers, like, like a few thousand people using it. And we're like, like, this has just blown our minds, right? We're like, this is kind of crazy that it just spread. And all of that was purely through the product. It was just like we didn't, I would say, through the product. It wasn't like some clever viral flow or whatever. It was like one teacher using it or it.

saying this will solve a problem for me, and then spreading it to other teachers. And so we'd find one teacher in school would pick it up, and then it would, like, travel through the whole school. And that happened a lot in the early days. So that was kind of early stuff. It was kind of we weren't doing a huge amount on trying to drive growth. It was more just like building a good product. But a little bit later, probably the biggest turning point, and it's this thing that continues

to this day, was we went to observe a teacher. classroom and Bessie Carmichael Elementary here in the city. And her name is Jenna Klein. And so we went to sit in her classroom and we were like, wow, she's a pretty awesome teacher. She was a science teacher, right? Like she's a great classroom and whatever. And she was using Dojo. And we thought nothing more of it. We had a bunch of research notes, went back to the product team, whatever. Two months later, I get an email from Jenna. And she was like, hey, I'm thinking of transitioning out of the classroom. and I'd love to come and do an internship with you guys. Like, would you be up for that?

And, you know, we were like six people at the time or something. We're like, I don't know what we would do with an intern. You know, like, and we're like, sure, like, we've probably got no, like, support or something that needs doing. Like, you can come and help out. And so she joined for the summer. And she was doing a bunch of, like, support stuff. And at the end of the summer, we're like, well, I guess she's going back to teaching now.

And she came to me and said, Sam, I think we're really missing a trick with community. And I was like, what do you mean in community? Like, we're building products, right? And she was like, no, no, no, like, I found this thing where, like, you'll get one teacher in a school who loves this and then spreads it to other teachers. And it's a real pattern. And here's how, like, the growth rate of, like, she done basically a whole analysis of, like, how schools were they with an influencer teacher who is spreading the word grow faster than other schools. and she'd got these like, like, she'd set up like a Facebook group for a bunch of these power.

She'd done all this like secret work basically in the summer on community stuff. And I was like, yeah, I guess we can extend the internship for a few months and just see how it goes. But this was like one of the big secrets in the company was like teachers are like amongst the most underserved, I think, people in the world. Like if you think the average teacher, they get into this job, which is. like, you know, pretty underpaid relative to most other jobs you could do. It's, like, often underappreciated. It's really tough.

And you're, like, on your feet with, like, 30 kids all day. Like, that's a really tough job. And so if you can, if you turn up and you're like, hey, we're going to be the people who, like, really listen to you and help you change your classroom to be the one that you dreamed off around the, you know, rather than, like, thing. Like, like, you've got your hands tied the entire time. It turns out they tell people about that. And Jenna built this crazy, like, like, you know. not crazy, I shouldn't say that, but it's like a community of teachers who are just so enthusiastic about changing their classrooms and helping other teachers change their classrooms.

Just making these connections among your users. Yeah, I was going to say, can you explain how that dynamic actually works? Yeah, yeah. So you say, you're looking for a teacher who will be an advocate in any particular school, right? So how are you finding them and then what happens from there? Yeah, so usually they'll find class dojo and then we'll support them to like spread the word bit more. So Jenna made this community called the mentor community.

And the mentors are basically our first teacher in every school that we're in who finds class dojo. So we still don't do that much to spread it from new schools. Like people, it jumps through like social networks. It's still primarily organic. It's still primarily organic. We still don't pay for user acquisition. So it's all like word of mouth. But once it gets to a school, then we really support that mentor to spread the word and to bring other teachers along if they think it's good.

Like we don't want, we don't want like a top-down implementation. We don't want the principle to grab it and push it onto everyone. We really want it to spread like in a grassroots way. And so Jenna built this whole like community, which is like supportive of each other and supportive of like spreading the word. And that's like a really, it's a power. It's a community with a purpose,

which you class as a movement, and that's like a really powerful thing. And so I think the turning point versus Jenna. And like if strategy one was just like good product, obsessively listening to people and serving them, Strategy two then was building a supportive community that is excited about like a different future for education. And I think that's like a secret. The community stuff.

I don't think many companies really understand that. Did Jenna stay with you? Yeah, she's still with us today. Nice. Yeah, she's awesome. Good job, Jenna. Yeah. That's awesome.

So Jenna's with you and you're more than six people, but you're still a small team. Still pretty small team. And class dojo. So can you give us a sense of the. reach your reach now in terms of the product. Yeah. Yeah. We usually talk about the percentage of schools that we're in.

So in the U.S., we're in like 90% of K-8 schools in the country now, which is about 100,000 schools or so. Okay. You're in about 100,000 schools? About 100,000 in the years. Okay. And so there may be one teacher. Maybe one, there's more and more, it's like whole schools, but it's at least one being active.

Okay. And then there's actually the first year that we're bigger internationally, in the U.S. So more schools outside the U.S. now using it then in the U.S. So more than 200,000 schools using it then. Okay. Got it.

And about half those in the U.S. So you're used by all of these schools and you guys have raised over $30 million in funding, but you're still a team of just 30. And, well, is that unusual given where you are? And was it a conscious decision to kind of keep the team at this size? and Raleen. Can you tell us about that thinking and kind of how you make that work if that's unusual?

Yeah. I mean, judging by the way you're asking it, it sounds pretty unusual. I think so? I don't know. I'm like, is this the right question? Yeah, absolutely. Yeah, I think it's... Like, the average company would be much bigger in terms of headcount.

Yeah. Yeah. Yeah, I know it's something we've thought a lot about. It's definitely intentional. there's a few things for me. One is that, like, I think it's always surprising how much small teams can get done. And I think there's actually, like, large diminishing returns, like, very, very large teams. And so for Dojo, like, we've kind of been informed by, there's a great book called Team of Teams on this.

I keep plugging this book everywhere. But, yeah. But, yeah, it's basically, like, how you create. I'm really interesting this as a, as a, as a, you're. as an area of personal interest is how you create high-performing teams. And this book talks a lot about two concepts, like empowered execution and shared context. And so I think the conventional way of, or a conventional way of building teams is that you hire like the VP who then like issues the instructions to the troops and the troops go and execute.

And I think in like fast moving or uncertain environments, it's far more interesting to have like a team of leaders who are capable of making decisions quickly and moving. quickly. So I think what we've done instead of building very large teams is build like, like the 30 people are structured in like even smaller teams, right? Like we have product development teams which are anywhere between like three to 10 people, a mix of engineers, PMs, designers, etc. And so we we try and have each of the teams having like a clear mission that they all believe in and are empowered to execute on it. So they've got all the information they need. So internally in the company, everything is public. Like we board decks are

public, financials are public, you have all the information to make a globally optimal decision about what you should do next, rather than just getting a sliver of information and having to march forward on that. And so I think that's actually how you get the most from people is when people are really bought in on a mission they believe in and they feel like the owners of that mission. So I think we've obsessed a lot over building that kind of culture. I don't think we always get it right, but we've seen great returns to really obsessing of like the right kind of culture. So we could certainly do with many more people on the team. Like I'm not saying that this is like, you know, this is where we'll stay now.

In fact, our first recruiter just joined last week. So I've done all the recruiting so far. But I do think the small team thing is like really undervalue. I think if you can get small high-performing teams working, you can scale those a lot. And there's actually something that Amazon talked about a lot with like the two pizza teams and things is if you can get small autonomous teams. What is the two pizza team? Oh, like the whole team could be fed with two pizzas. Oh, I see.

Yeah. Okay. I can eat two pizzas. So that's why I do all the podcasts on my own. So what would you advise people who want to adopt that same strategy, but they're also used to doing all the stuff? They're like, I want to do all the recruiting because so far I've recruited 30 great people. Yeah.

And how do you like, how do you train yourself to let go? Yeah, I think so specifically for me as CEO, I think it's the, there's this great thing in Dalia's book principles of what your job becomes. And he's got this diagram which is based like goals, machine results. Okay. And the machine is just people in culture. And like from periodically, you kind of want to step back and think about like the design of the machine and like what your role is in the machine. And Ben, uh, Ben, uh, Ben, Ed Evans actually has a good post on this. It's about Amazon. He calls it the machine that builds the machine. And, um, I loved it. But I think this is like a, it's an important thing to be able to get some distance and reflect on what the right,

what the appropriate role for you in the company is. I think actually Sam Altman said this, that you get to about 25 people and your job shifts from building the product to building the culture and it kind of stays there. And that's so true, right? Like the reason I've just replaced myself with an actual recruiter is that that's not the right job for me anymore. There are other parts of the culture that I need to focus on and build. So I think it's, to answer your question, I think it's,

having a retrospective loop on what your role as CEO is in, but designing the whole machine and then your part in that machine. Okay. What other books have been helpful for you in developing a leadership in product strategy? Oh, there's a lot. There's, um, we have a pretty obsessive reading culture at a dojo. We have a book club every month and like, like, most people like do, I think, an hour or so of reading a day at least, but yeah, it's just quite good.

But there's been a few. So I've talked about team of teams. There's extreme ownership is another good one. There's kind of like the convention, like the zero to one and all that. Like the startup e-books. The startup-y kind of books. Yeah, that everyone will point to me. I'm trying to, I should pull up my Kindle and think of what some of the unusual ones are.

But I really love principles when you read it a while ago. Yeah. It's certainly a little bit of extreme, but I think it's quite good. There's one called the Evolution of Everything. which, like, the main message from it was that I think it's very easy to believe the world is, like, designed in some way and embracing, like, emergence in the company. Like, for instance, the growth mindset videos I spoke about. Like, that was an emergent idea. That wasn't CEO said we have to do growth mindset videos. That was, like, one of our product development teams, like, came up with this insight. And now that's, like, a major part of what we do at Dojo. And so that embracing the idea of emergence is a really, really important moment. And do you draw inspiration from books that aren't about business? Yeah, I mean, there's a lot.

There's one that we at the moment called The Diamond Age, which seems quite good. Okay. Yeah, I'm quite early on it. But I'm the ones that are coming to mind for me now with all the ones I've read recently. Yeah. That's everything. Yeah. Yeah. Karen, I have a quick.

What was Sam like in the early day? What do you remember about him? In the early days at Imagine Key 12? Sure. Yeah, because we met when you moved to America to build this startup, but you didn't know what this startup was yet. It's just valuable because so many people listen to the podcast or watch videos, even like the authors you're talking about, right?

Yeah. Oh, this is Jocko's so hardcore. He's so hard to become Jock. What's the workout plan? Yeah. There's just some guy, right? Yeah. He was an 18-year-old at some point, and the same is true for all of us, right?

So I think that perspective would be interesting if you remember it. Early days of Sam, I remember, I mean, you're very easy to talk. Sam is very easy to talk to. And so that kind of stands out and I think was probably part of what made that early learning from customers so possible. But yeah, I remember, you know, those early months, it seemed like a very stressful time because you didn't know exactly what demo day was coming. Product didn't exist yet. The idea was taking form. But Sam and Liam would just kind of show up and be like,

like, yeah, we're working. We're getting all this, you know, we're talking to all these people. And we're kind of wondering, what's going to come of this? What's going to come of this pet? And here came glass dough. And I have this vivid memory of Sam. You're talking about your frugal British nature. Sam wearing glasses with only one side of the glasses was still holding them to his face.

And the other side was missing. Is there a word for that, like the love of like patching things together in the UK? You see it with houses, too. Yeah. You know, there's, like, the idea of, like, the crazy, like, British inventor. Yeah, yeah. There's, like, the one person in a shadow ever. I think that's actually Liam.

That's not me. But, like, that's, I think that's, uh, there should be a word for it. Yeah. Yeah. Well, and it wasn't quite patched. It was just dangling on the face. There's not no patching. The glasses still seem to serve their purpose.

Yeah. You don't need to. Those legs. Yeah. Really, really proud of days. And what has changed since you guys just start? Like, I'm sure you had it. some degree of vision, right, in the beginning and what you thought the product would be?

What, what changed? What were some big realizations? Oh, man. Yeah, that's a, that's a big question, too. What's been amazing is, like, the core thesis that I said at the start that, like, I actually think education is made by people, not by technology, and the role of technology is in service of people, supporting them. That thesis hasn't changed, like, from those early conversations, it became pretty clear

that the best way we could have impact was by serving teachers, helping them connect with the kids and parents, and then helping them improve their classrooms. And that would be a way to both have impact and to scale. But I think the thing that's changed the most is the actual products that we've built. So class has never been like a one product company. I think like the first problem we ever worked on

was helping teachers build a better classroom culture. And that was just a product for teachers to use. inside the classroom. Later on, it became far more important for us to share beyond the classroom to include the whole community, families, also other teachers and the principal in the school. And so it became much more of a network product than just like a single player kind of thing. And then the thing that changed after that was that network could then be used to share more and more ideas, right?

And so that was probably the latest evolution. So it's kind of like the product, I think, has changed a lot. I think the thesis for the company stayed. largely the same. Okay. And was there, were there learnings that may be counterintuitive that are maybe, I guess it could be for ed tech specifically or just a startup in general. I'm always curious to see what people pick up along the way that they didn't expect at all in the beginning, that you would, you would advise other founders to keep in like. Yeah, I actually wrote some of these down. I mean, just put one of these up. Yeah, I thought about this one, because I saw it. Yeah, I think like what

one the ones from me is this thing of this we talked about a bit but like the small teams thing because the whole way we've always been pushed by investors to like just like you've got to like hire more people and grow faster and whatever and you think about this a bit from from the realities of the education world like you work in school years so there's no point you're being alive for like three out of the 12 months of the school year you better be around in 12 months like chunks and so I don't know if that's counterintuitive but it's not obvious from the outside. You'd be like, well, we should just hire people in like January and then, you know,

it's like, cool, but if that means you run out of money in June, then you don't get another school year. Yeah. So that was good. And I think like, it's been surprising. It's always surprising me how much small teams can do. So that was a big one for me. Another one for me was like the, the emergence thing.

I think there's a narrative in Silicon Valley that, like, you know, some kind of, you know, some kind a far forward-looking visionary will like come down from the mountain with the plan and then give it to the troops and like the troops will just go and execute it and it may be true right like there may be people who are just that brilliant and like that that's I think that is the case in a few instances but for us I think we've got a great um we've got a lot of mileage out of just like embracing the idea of like your job isn't to like minimize the chaos but it's like empower people to think big and to be bold about the things that they might do. And the only way you can really do that is one by making the company a psychologically safe place. And the teams

a psychologically safe place. Like if you're always being judged on like success and failure at every moment, it's really tough for people to think big. How do you do that? I mean, because that's an issue everywhere. Yeah. So one actually interesting thing we did recently, it wasn't so much for products, but it was across the whole organization, uh, was we started a running log of all issues in the company. Okay. So this is really weird, right?

And it's internally public for everyone to see. And anyone can add to it. You can add anonymously. You can add like, like with your name on it. Um, and so it'll be everything from like, hey, like the priorities aren't clear to, hey, like, we're really like creating too much trash to, um, what happened with that, you know, that X project that we never heard about. Yeah.

And so the idea is that, like, I think we really value building a culture where we can be, like, real with each other. And it's never in, like, a hard or arrogant way. It's actually in, like, the most empathic way that, like, we all think this mission is, like, really, really important. It's really important that, like, every kid in the world gets a better education in a reasonable period of time. And so we're all here to do that. And so any issue that's standing in the way of doing that, like, it's far better to, like, drag it out into the light where it can be solved than to, like, have it fester quietly. And so I think that, like, building the kind of company that, like, quickly surfaces

and resolves issues rather than letting them fester is really important. So, Mike, like, for that one practice, it's recent, but, you know, we review that log every week. The guarantee is that there will always be action taken on it. Like, we'll prioritize it and we'll take action on it. But you can, we do it in Asana and you can, like, follow along. So if you submit an issue, you know it's going to get worked on it. The idea is not a new one.

It actually came from, there's a plant called the Numi plant, which was, the partnership between Toyota and GM, and it went from one of the worst car production factories in Fremont to one the best in the world. And it did it by really empowering the people doing the work. And they would guarantee that any suggestion you put in the suggestion box would be acted on straight away. And if it couldn't be act on for some reason, they would publicly announce it within a week, why it couldn't be acted on. And so it's like, you imagine what you imagine a factory looks like. It's not that. It's like you just do your little slice of the assembly line and move on. Any of, I'm digressing a bit, but I think that was really counterintuitive

that, like the standard narratives about what great companies and great leaders do, which is, you know, set the vision and tell people what to do is like, we're so far off for us. And there's strong parallels with classrooms. I don't think, I think great teachers don't tell the kids what to do. Yeah. And so it's a nice, like, golden thread from inside the company to out of it. That's a great insight. I used to work at the onion. And when everyone, it's something that's not necessarily obvious, but the onion doesn't have bylines. And that anonymity is amazing for creativity. Wow. Yeah. Because it's so easy. I don't know. This might be a little racy or this is like super weird. Like I don't know if I want my name behind this headline. Right. It doesn't matter.

It's all by the onion or it's by made up people. Right. It's like a non-bye. Yeah. Yeah. And it was so handy. And so yeah, to your point, like I think if people could submit, you know, issues anonymously, but also see it tracked. Yeah. Super helpful, especially, I mean, place like YC, same thing. Yeah. Like, we could do a better job. Yeah.

So your product serves a, it's like in a sensitive space. You have sensitive audience, right? You're used in schools by teachers teaching young children and everyone's very interested in what young children are touching and what kind of lessons. and what kind of lessons they're receiving. And a few years ago, you went through some kind of court of public opinion trials. So I want you to share about that because I think, you know, every company has to worry about how they're perceived and how,

and that court of public opinion. But it's really amplified when you're dealing with an audience that includes, when you're serving customers that include young children. So I want to see if you could share. a little bit about that story and what you learn from that kind of as a company in terms of how you manage that side of what you have to be. Yeah, no, no, totally happy to share. So I think like this actually, I think Imagine K-12 had an amazing role of playing this.

No, no, no, no, in a bad way. So like five or six years ago, right, like we, there was this new cohort of like ed tech companies that just like launched into the world. And it was a really, like, uncertain world for, like, all of us. I mean, you know, you guys were also building Imagine K-12. And, like, there's lots of learnings there. There are lots of learnings for us. So we did, like, the conventional thing, which was, like, build a product that people love, get it out, lots of people.

Focus on that. And all of us did it because that was what we were, you know, that was the conventional wisdom of the Valley at the time. And I think it was, like, a few years in, I forget exactly which year it was. But there was really, like, this sudden, like, oh, my God. Like, there's all this technology in classrooms now. Like, what is it, is it good? Is it bad? What is it?

You know, et cetera, et cetera. Like lots and lots of debate and discussion about it. And it really, like, dragged a few things out into the light for me, right? Because, like, the naive way to build products is, like, we'll just build a great product and everything else will take care of itself. And you most often hear that with, like, founders who don't want to do sales or, for instance, right? But this one wasn't sales at all. It was actually, like, clear communication. And so as a company, which, like, you know, we have a lot of introverts.

in the company and I used to be very expert, but more commarginally over time. Like, you kind of forget that, like, that's actually, like, really, really important. Like, it's important to do what you say and so what you do and to be clear in your communication. And so you'd had, like, a few years ago, you had, you know, a bunch of young startups who had suddenly got to scale or some scale, interesting enough scale, that it was relevant, who had, like, boilerplate, like, terms and conditions or privacy policies or hadn't updated the website, site, and we're guilty of all of these, but I had an update the website to, like, to actually describe what you do.

Or had cater to one audience like teachers, but not to, like, districts. And so, like, I think the thing that, and we actually met, the thing that, like, started to solve it in the end for us. What was the, what was the criticism that you guys were dealing with? The criticism was, like, yeah, no, there are lots of criticisms for, I think, not just for us, the whole industry, but for us, it was like, well, like, what happens with, like, kids information, right? Because the only parallel people have,

and there's actually two points here. One is communication. The other is like, what's your business model? Right? And so the first one in communication is important for everyone. The second one, if you go back a few years now, the main example people had was like, business on the internet was like Facebook or Google.

Like these are the companies often get held up, right? And they're like, oh, like those companies like literally take users information and give it to advertisers. And so when companies are working in classrooms, I don't think that's acceptable at all. Like, that's not, like, that would be a really bad way to build and fund an education technology company. And so I think it's really, so in the first,

it's important to be clear and crisp in your communication, and I think many of us weren't there. Like, we, you know, I know a few of us at least wrote our, rewrote all of our policies in like sixth grade language or sixth grade level English, just so that anyone can understand it, right? Because legal job... Your terms of service.

Yeah, because legal, legal jogging is legal jargon, and it's confusing and weird for everyone, right? But if you write it, like, what does this mean? That, like, that goes a long way. Also being clear, like, hey, this isn't an advertising, like, a company, your data's not going anymore. It's yours.

You can delete it. You can change it. Like, if you don't delete it, we'll delete it after, like, X years of inactivity or something, right? Because we don't need it. Like, it's yours. Like, it's not... We don't need it to sell to ads.

sell ads against. And so then the second thing was like making sure your incentives are aligned with your users. And so for us, like actually the best business model in the world is like also the oldest one in the world, which is like, hey, we'll make some stuff for you. And if you really like it, you can buy some of it, you know, like which isn't like exercise, right? But like that's all we're saying. We're like, hey, we'll make a bunch of stuff for teachers and for parents. And we're particularly interested for parents. We're like, well, most parents don't really have very good choices. Like most parents in the world, you have the choice of if you want to do something good for your kids' education, if you're in the top 10% of the population, you can send them to private school and pay $30,000 a year.

But if you're not, like, what do you do? And we're like, well, if we could give you a better choice to make there, which is also affordable, et cetera, we give you, we call it the education bundle for your kids. Like, would that be interesting? And parents are like, yeah, that would be really interesting. And so I think just being clear about these things is something that we weren't guilty of because we're so heads, we're guilty of because we're so heads down like just building products because you know make something people want right sure and you know your intentions but right but what else you say

and and i think it's one of those things of like leveling up as a company where you realize like the scope of your responsibility extends beyond like your own user base and beyond like your team and beyond um you know to people in classrooms who we are obsessed with and love it extends to people who have never touched your product or used it and they need to know too and that was like a whole different thing. We hired a head of comms that year, and she's still with us, and she's amazing. But that really built a new function in the company, right, that we didn't have. As you expand internationally, as you said, you're now used in more schools internationally than in the U.S., is that kind of changing the communication needs? I mean, there are different

kind of privacy regimes in the EU and so forth. Yeah, what do you see? Yeah, I mean, the short answer is yes. Sure. like you do the right thing for every jurisdiction you operate in. Anything kind of interesting or your counterintuitive coming out of that or it's pretty straightforward. You just kind of apply by that. I mean, it's pretty straightforward. The rules are all different, but they're all in the same spirit, right?

But in terms of communicating with parent communities so that, or maybe, I don't know, maybe penetration in those different countries isn't at the level where you're getting that scrutiny kind of in the press and so on. Maybe it is. I don't know. Yeah, no, I mean, like I said, We're majority international now. So the growth is good.

But I think, like, the EU is, like, kind of famous for, you know, being very buttoned up on privacy. So there's a while ago there was, like, I forget it's called now, but it was like the safe harbor thing. And then there was the privacy shield. And so you just kind of have, we have, like, people in the company who, like, full time stay on top of that stuff. Sure.

Which you don't have to think about when you're, like, and, you know, with AT teachers. So I think just as, like, as the company grows, you have, like, more of these considerations to take into account when you're building products. Like our products all go to, then we have like counsel who like looks them and makes sure they're compliant with the different jurisdictions. But as far as managing the public conversation about it, has anything interesting come up there that was different?

Not, not obvious thing that comes to mind. I'm racking my brains think of one. Oh, that's good. Yeah, no, no, not obvious one. I mean, like, I think in general just like being clearer in communication in all of our products, like helps everyone, right? It's not like so scoped to jurisdictions. If like if every parent is like, oh, cool, like I get to, you know, share in my kids' school day now.

Wonderful. And their classroom is getting better. Awesome. You know, but we never say those things before. So, yeah. You mentioned charging parents earlier for certain things. What does monetization look like for you guys in the future? Yeah.

I think we've always had this is one of the things that hasn't changed from the start, actually. Yeah. It's like we've had the same view on it that. So a few principles. Like we really do want to reach like every classroom in the world. Like, it'd be really good if we, like, it's one of the, one the flaws with the current education system is that, like, I've already talked about one, just that, like, there's often stagnation. You can't get new ideas implemented quickly. It's all like, what would you call them, like, type one decisions, right? They're like big heavyweight kind of like, do we do this or not? And if you can make it like a type two decision, make it really easy to try and reverse, like, wonderful. I think that that's more innovation happen. But a second flaw in it is, is the inequality. You know, people will talk about how education the U.S. is broken. Like, like, it's not actually for a few people. It's really. It's really. really good for a few people. And for most people, like, the difference between the luckiest and the average is, like, large. And so for me, like, the, if you're working in education,

there's almost like an imperative to work on the, the equity part of it. And so we don't want to do anything that's, that prohibits us or, like, gets in the way of that equity argument. So we do want to get to every class in the world. On the other side, I think, like, the business models education have mostly been about selling to schools and districts. And, um, and, this is like slow it takes a while it you know it would get in the way of that that that goal um but there's a really interesting observation we had that i think i've already alluded to that like parents like everywhere in the world like for the most part really care about their kids yep it's something we've done for like a long time right for like for it's like it's like it's an evolutionary

thing it's not just a job it's just like we're wired to care about our kids you have to keep it alive like the small ones you very clearly have to keep alive yeah yeah it's a tiny tiny tiny person that depends on you. And so we're like, well, these are also people who are like really underserved in my view. I'm like, like I said, you, you have fairly limited choices. You hope the school near you is really good and that the specific teacher whose classroom you go to is like, that's a really good teacher. And maybe if you can afford it, you like move houses or move jobs, whatever to get to a better school. But like, you know, for most of the world's population, that's pretty unreasonable. Maybe you can go to private school if you can pay the fees. But like,

again, pretty unreasonable, pretty out of reach. And I'm like, well, if we could just create value for parents, enough value that, like, there's something that they want to pay for, that could be interesting. And if it could be, like, relatively affordable on, you know, the order of 10 bucks, let's say. Sure. Like, a month. Like, that would be amazing. And so that's kind of the general direction we've gone. And we've, and that's been the same direction that we've pursued from the start. So, yeah, another common question that comes in over the board a lot at YC is How is ed tech different than traditional or whatever startups, tech startups? And actually, probably both of you guys could answer this one.

Yeah. Do you want to go for? Yeah. Oh, that's been nice. There's like the obvious differences I think that we've spoken about around like, hey, how distribution is different. And like things around privacy and security need to be buttoned up on and how it's not quite consumer or not quite enterprise.

Yeah. But I think the non-obvious one for me is how. education isn't it's not a tech product it's and it's really like because like unlike other industries I think education is really just what happens between a teacher and a learner and so it's really like a very human creation and so the role of technology is is really like to support that not to like replace it and the analogy I was used for the team is like if you and I think this this is really important because like the analogy I use for the team is a difference between a technical system and a human system.

And if you're making like one perfect technical system, like a MacBook or a Tesla, let's say, probably the right strategy is to have like a perfect design for it and then just like sell it to as many people as can afford it and hope it gets cheaper over time. Like it's like design first and then give it and then sell it to people. But I think if you're working with a human system, like a team or a classroom,

like it's actually a really bad strategy to like design something and then just like impose it on people. I think you actually fall the opposite order where you get two people and you help them design the changes they want to make in their human systems. I think that's like a non-obvious but really important mental model difference. Would you agree? I haven't thought about it in those terms, but absolutely. Like you said, this is a human system and you need to be worried about, not worried about, but conscious of the fact that it's going to have, it's not going to be consistent across every implementation because you have humans evolved.

I would also say ed tech is, you know, has a very compelling double bottom line. You know, companies can get into this because they can build interesting businesses, but they can also have a big impact on this equity problem. That's very interesting and compelling to work on. It's also right for lots of solutions. to come in and make things more efficient. I think if we look at kind of the administrative level in education systems, in K through 12, especially in districts,

they haven't been well served by enterprise products that, you know, they try to adapt into the education system or buy a lot of the legacy education products. So we see a lot of great companies being built around, not being so focused on the classroom, But in this back office, how do we do these things efficiently that we've figured out how to do well in enterprise settings? But something needs to be a little bit different in a school setting. And so there's just a lot of room to build interesting companies in this space still.

So it's an exciting place to be working in. Yeah, I think that's great. All right. Well, thanks for coming in, Sam. Cheers. Thank you. Yeah, good to see you. Yeah.

Take care. I appreciate it. All right, thanks for listening. So as always, you can check out the transcript. video at blog.w.Ycombinator.com. And if you have some time, please leave us a rating and review wherever you find your podcasts. See you next time.