Keith Schacht and Doug Peltz on What Traction Feels Like - at YC Edtech Night

This is the last fireside chat tonight, and I am very happy to introduce Quiche Act and Doug Pelts from Mystery Science. Thank you! Whoo! Thanks! Could you guys just start us off by introducing yourselves, please?

I'll let you go first.

Okay! I'm Doug from Mystery Science, and I taught elementary and middle school science for seven years before this guy convinced me to quit my job and join a start-up in San Francisco.

And I'm Keith. Doug and I have been friends a long time, 16, 17 years; we lose track. We met in college, and Doug went right into the classroom. Outside of school, I started a couple, three startups and had a certain small to medium level of success, and then joined Facebook. I left Facebook to start Mystery Science and convinced Doug to come along with me.

So, just to kick things off, what is Mystery Science?

Good question, Jeff! That was a softball right there. So, Mystery Science, from the teacher's perspective, most elementary teachers in this country—kindergarten through fifth grade—about 95% of elementary schools in the country have zero science teachers on staff. So elementary teachers are expected to teach science, and it's a subject they struggle with the most.

So, Mystery Science is an interactive curriculum that they now use instead of their textbooks to help them teach science. We often describe it as feeling to them like a virtual science expert named Doug who co-teaches the class with them remotely. It's a sort of pre-recorded, choreographed experience.

How many schools teach Mystery Science?

It's always a fuzzy number to count. On a weekly basis, we have active teachers now in more than 10% of schools. Well, it might be a lot more than that now; I forget.

That's amazing! You gave a little hint at it, but tell us the founding story of Mystery Science. It sounds like someone had to be convinced. Sounds like there was an inspiration. Can you sort of walk us through how it happened, how you got together, and how you actually kicked off this crazy venture?

Well, Keith, as he mentioned, we met in college, and he was always my tech engineering friend. He would explain to me, like I’d say, how does email work? Like, what happens when I... I would use what happens when you click "like" on Facebook? But we're old enough that there wasn't Facebook yet, so I’d be lying. But he would explain that to me.

Yeah, and Doug was my nature friend. He would be like, I’m taking you out of the computer lab, and we're going on a hike, and you're gonna walk through the woods, and I gotta show you cool stuff. So we always had this fun back-and-forth. I knew once I got to college, like, I had always loved science from the time I was a kid, and the only career path is, you know, you become a scientist; that's what everyone told you. I never had a—it's so funny, so many of my friends who work in education cite, like, an influential teacher. I never had, like, a great science teacher. I liked science in spite of it. I figured that when I got to college, science explanation would be so much better because now the scientists themselves would be teaching you, and it turns out they're even worse at explaining things.

So I got really interested in the problem of explaining science, and I figured out while I was in college that I wanted to become a teacher and write my own curriculum. Keith went on a little bit of a different path, but it was around the time that he had kids that he started taking an interest in my career. We always had this shared interest in explaining science and technology to other people.

Right out of school, I did the same thing; I did it at an advanced level, helping R&D engineers learn about new materials and manufacturing processes and electronic components and things like that. Doug and I would have this fun exchange every year. We’d get together, often over Thanksgiving, and teach each other things. As he mentioned, I had kids. I was at Facebook. My wife and I were visiting Doug and his wife over Thanksgiving, and you know, once you have kids, you start thinking like where am I gonna send them and how am I gonna be responsible for this other human being's life for the next of whatever, 12, 16, 18 years?

I saw Doug teach in the classroom, and I was very interested in education. My wife was an educator as well, and seeing Doug teach in the classroom, I thought, okay, my kid is gonna learn science from Doug—like, check! Alright! I don’t want to move down to Orange County, but that's somehow a piece of the equation. That series of conversations started then of, like, okay, we should—we have this shared passion; we should really work on this.

And I’ll skip a few steps in the story, but the passion, the sort of shared passion we had was that there's this fascinating body of knowledge in the world about how the world works. This is what scientific technology is; it's what makes our modern lives possible. We all learn about it kind of in school; we take science class, but it's very often one of the most boring classes that you take.

Like, it is labeling the layers of the earth: across the mantle, there’s a molten core in the middle. You guys remember doing that? And you label the parts of a cell; there’s the mitochondria—there’s color, those like purple. Yeah, and that’s what science class has become; it's this exercise in vocabulary memorization. But it wasn’t that for us.

So that was the spark of the idea, was like there is this huge disconnect here between how fascinating this content is and how terrible it's being presented. We should do something about that. Now, it was not obvious that we were gonna be in schools and create a curriculum and all that stuff, but that was enough that somehow convinced his wife to leave his job and move up here—somehow he handled that part.

So when you started, you didn't even know who your customer was. You didn't know you were gonna sell to schools; you didn't know what you were gonna sell; you just said we’re going to find a better way to teach science?

Yes, yes! I mean, we knew our customer was children. So, you know, I now think of us not as an education company but as a learning company. Like, we almost fought kicking and screaming into schools, you know? We knew we wanted to reach children, and parents, we assumed, were our gatekeepers to children. You can't get a six-year-old to sign up for your website and start using something. So we’re thinking about children five to ten years old, and we prototyped a whole series of products initially for parents, and then some friends of ours were homeschooling, and one of them happened to be for elementary teachers, but it was with much reservation that we started thinking about schools.

Ben Horowitz spoke in his book, called The Hard Thing About Hard Things, and it sounds like, you know, if you just step back a second to think, we're gonna reinvent science curriculum for kids, and we’re gonna make a business out of that—that sounds... did you know how stupid and hard that was gonna be?

Halfway through, you know, we managed to convince some people to give us money early on, you know, in spite of it being stupid and hard.

Yes! I mean, I think we had partly—I had done this a few times before, so I expected, you know, I said to Doug, I'm like, well, the last three times I started something took me six to eight months to go from a vague sentence on a napkin to, like, you know, a meaningful business. It took us more like 18 months, so it was harder than we thought. But I told Doug it'll take six months; don’t worry. I remember six months... you were—what did you say to yourself when you were still 12 months away?

Maybe six more?

I mean, that was—I can... Though, so then in the first year, we probably went through—depending on how you count it— we probably went through five or six different, like, products. One of them, we spent almost six months on, and we thought that was the one.

Yeah. For all... we almost thought that the... I thought it was the one longer.

What was that product?

Basically, that was like Pinterest for science teachers. It was really dumb!

I like to give it a little bit more credit; it was the resource Doug wished he had, was a science team.

Oh, yeah! I built something for myself!

We just started; I loved it!

How did you find out that it was the wrong idea?

Well, so here's our thing earlier to what Tim was saying about teachers: they are nice to a fault because we'd get them on the phone, and they weren’t just nice—they'd be like, this is so interesting! I like—and especially when I would explain it. I’m explaining, like, my vision for, like, imagine a lesson that's all these visuals.

And so the—where the rubber hits the road is when you start watching the usage. I mean, it’s like they’d say it’s so interesting and then it was just like that.

So we reached a point where, I remember, Keith said, like, it was heartbreaking to me—he’s like, we’re going to pivot again. Again! This is wrong! And I was like, no, no, no, no, we just gotta give it a few more weeks!

No, no, this is wrong!

You remember this moment?

Because you will know what traction feels like.

Yeah, you're saying that to them, or he said that to you?

I mean, because there was this—you know, it was not an argument, but there was a disagreement. He was like, I think people like this! Like, I wanted this... Look! We talked to the think of those 10 teachers in the last two weeks, we talked to—they all said this is awesome! And I was, you know, I remember saying to Doug, this isn't what traction feels like. Like, you will know when we have traction, and this is not traction.

By the way, this is one of the most non-intuitive things about—we talked about knowing when you have product-market fit; sometimes it's not so obvious, especially if you build an awesome product that it turns out customers didn't want to use.

What happened that was interesting was we thought—because I was a science teacher, I was a science specialist—so we thought we were going after science specialists. And I’d explain this to them, and it was a little disappointing that they wouldn't use it. Like, they didn't do science in the same way. In hindsight, this is obvious! Like, they don't teach science the way I was teaching it.

That's—we're doing something unique with how we're explaining science. What happened was we would accidentally get an elementary school teacher in the funnel, and I remember the first couple times that happened because I was a little annoyed because I’d look at Keith and be like, uh-oh! This is a third-grade teacher; this isn't the person we want to talk to, right? And we posted on the middle school science teacher this way; somehow we got the third-grade...

But we’d explain, oh, so here’s all these pictures, and the third-grade teacher would say, so what do you do with these pictures? And I’d say, well, let me just walk you through how I would use them. And then it was like, after three times that that happened where a teacher said, I wish I could have just recorded everything you just did and play that back for my kids; that’s when we were like, huh, I think we solved someone's problem right now.

So it wasn't exactly sort of a sophisticated product segmentation analysis that got you to figure out what product-market fit was; your users walked you into it.

Yeah, I mean, it was systematic about, this is what we know, and this is what we don’t know, and okay, well, if that’s not true, then what else could we? And so it was clear that, you know, to connect the dots a little bit, Doug had this unique way of teaching science, the key ingredient being that he actually explains the evidence for how we know the conclusions.

Like, funny! I don’t—you know, like, you know there’s a molten core in the middle of the earth, right? Like we all learned that, of course, we know there's one, even though no teacher ever told us how humans have figured out there’s a molten core thousands of miles beneath our feet. You know, that’s the disconnect in science class, so Doug teaches science class very differently.

Pictures play a key role because when you're presenting evidence, you often show the evidence for what it is you're teaching. And so this was a resource that was a collection of pictures. But middle school science teachers don’t teach science that way, so they were fascinated by it, liked it, but they didn't know what to do with it. And so the more Doug showed them, the more they started to look for finished lessons, and then elementary teachers, who got excited about finished lessons, they don’t have an opinion on how science should be taught.

And they were actually, on the contrary, it’s the subject they’re the most scared of teaching, so they're like if you could just step in and help me, that would be great!

So walk us through it a little more detail. What happened when you finally figured out, oh, this is our product, and you know, when you had that moment when Keith looked back at you and said, see, this is traction?

Yeah! I mean, it was like—and we've saved all these screenshots because we now walk all new employees through the origin story. And so there was a, you know, a screenshot with all these thumbnails that looks like Pinterest, and then they're ordered, and then there’s captions beneath them, and then there's an intro paragraph, and it’s like starts to look more and more like a lesson, and then it is a finished lesson.

And so there was a point where we said, okay, elementary teachers seem much more excited about this; let’s find 10 elementary teachers and 10 homeschoolers. So what we actually were considering both and get them all to agree to teach the same eight lessons in a row, and it was like December 1st, and we were like, yeah, so we’ll get all eight lessons done by the end of the year and then they’ll all start right after the holidays.

And then we had 20 people on this pilot, and this was, you know, pivot number 7 or so on our list, and we were pretty sure it was gonna work just like we were pretty sure the last six were gonna work. And it worked! They loved it! So, once a week, you know, starting January 1st of 2014, I think it was, once a week for the next eight weeks, we released the lesson and scrambled to get it done in time.

It was just a much—you know, they were incredibly enthusiastic about using it, and the feedback that we got—and they were sharing us anecdotes from the—sharing with us anecdotes from the kids. And it was clear it was very different. A lot of times you really know you have product-market fit and traction because it just kind of goes and happens, and all of a sudden you find people using your product that you had never heard of, and you know, your servers start to break, etc. How did that happen, and when did you know?

I mean, I remember one moment. This was back when—now in hindsight, if you want to put something—if you find a Facebook page to look for, it’s funny talking about The Hard Thing About Hard Things. You talked—he says don’t look for silver bullets. I love looking for silver bullets, though I ignore that advice. I still look for a silver bullet so that we could get the word out about Mystery Science to people. And I love using Facebook, so I would always look for, like, wow, there’s gotta be some Facebook page that all the teachers are on.

And I wanted to find this one that had, like, over a million likes, and it was this era when if you posted it, if you had a Facebook page and you had people like your page, then they saw your post. Now in hindsight—food! Yeah, in their feeds—in hindsight it was—the Facebook was giving you a free year. Like, looking back on it now, now it’s like, yeah, sure, you can give that to everyone if you pay Facebook a lot of money, right?

But we managed to convince this guy to put a—like, a little ad up about Mystery Science, and how many signups did we get in the day?

It was—we got about 15,000 signups!

Laughter. So your question—didn't you guys know that there were these huge gorillas that McGraw Hill, I don’t know, Random House, Pearson—who sells science curriculum to these schools? Didn’t you know you had no chance against them? Shouldn't you have known that this was a case of, you know, if you want to do something impossible, give it to someone who doesn’t know that it’s not impossible?

You know, it doesn’t know yet that it’s impossible, so we didn’t! You know, even at this time I was actually pretty ignorant of the whole curriculum company space, so I didn’t know that at all. Maybe you do that—we didn’t talk about that a lot. We—thought we were building a resource for teachers, and the distinction between supplemental resources and curricular resources was totally foreign to us.

And you know, one company in the space that served as a model for us was called BrainPOP, which is, you know, like a website that teachers pay money for that has a bunch of subscriptions. And you know, we researched all the sort of competitors in the space, and they claimed to have 20% of schools in America paying them, and they charged about $1,500 per school, and what we were making was not too dissimilar from that. And so that was one data point that we know it's possible for at least one company to pull this off.

And in retrospect, they're not a curricular resource, and so we assumed we were gonna be a supplemental... fairly early on, and so we didn’t really think about those companies.

Do you remember your very first sale?

No, it was kind of cheating! Like, the one thing we both had was we know a lot of people in the space—a friend of ours who ran a school agreed to pay us something on this side so that we could say we had a paying customer. I think that was our first.

Did you decide from the very beginning that this was a for-pay service? Charge for a while?

Yeah, when did we decide that? I mean, the ones—the one thing I can relate that might be of interest is I definitely remember our first user, like, our first teacher! We, like, enshrined her. I’ve joked with her, and she’s not too weirded out by this—that we should have, like, a cardboard cutout of her in the office because she—I tell, like, anyone that joins our company: you have to meet Kiyomi is her name and learn to think like Kiyomi.

Like, she is—we have different customers now, different mental models of like what our user is, but she is just the prototypical target user for us, and so I remember her. And then on the sales side, it was a couple of years in before we started sales, and I remember saying to ourselves, if we work really hard, this was in January, we were gonna start sales, and I thought if we worked really hard, we could sell 300 schools by June, and we sold 15.

So that was, yeah! What we knew fairly early on that we were gonna charge some kind of subscription fee for this. You know, as we started thinking about the teacher resource, but didn’t know how we were gonna sell to schools. I really didn’t want to build a sales team; we were just focused on growing for usage. So it was a freemium model, and we treated this like a consumer business, you know, a consumer web business: like teachers are our customers; let’s get them to spread us word-of-mouth, and we’ll figure out the home-making money thing at some point; that's how we held it.

So you guys just heard Doug just threw out a number, like $1,500—they’ve sold to a lot of schools! So you kind of have squared this circle; everyone knows how hard it is to sell to schools, but somehow you guys have managed it. Can you give a little bit of the secret sauce? If you're going to try to be successful selling into districts and schools, what do you need?

I know you should take the piece of the automation because the thing I would say, like, one piece of the sauce is when you're asking about Pearson earlier, right? It's like, yeah, those big behemoths, they make stuff and they sell stuff, but do teachers like it? And I hate naming, like, I’m fine trash-talking Pearson; I feel like everyone does that.

But, you know, there’s—I won’t, like, there are others—in the science curriculum space, let’s just say there are large boxes that have, like, an inch of dust on top of them in many schools, and ten years ago, they were sold to schools as a curriculum and teachers don’t use them.

So, the secret sauce for us was—we’ll make something teachers love, make something that the kids love, and if you do that, you solve so many problems because the teachers are willing to spread this virally via word-of-mouth and then even willing to be—not your salespeople, but they're willing to—when it comes time to, like, submit a purchase order to the principal, they’re more than happy to do that.

Yeah! Oh, that was not obvious, you know, early on. So, you know, we knew that we had something that teachers loved, and that we were, you know, fundamentally creating a lot of value for teachers and for students—and so that was sort of step one.

And we had done a pretty good job of growing. I think we’d passed a hundred thousand signups or so and so we felt pretty good about that, but it was non-obvious how to sell to schools. We hired—we, you know, we asked a lot of people for advice, and it was, you hire a VP of Sales, and you build a sales team, and so we hired a VP of Sales, and three months later, we fired a VP of Sales.

Was that the biggest thing that went wrong as you were trying to grow?

Probably! I mean, we—the hardest part of the business was—gosh, was it before or after that that we almost ran out of money?

That was after that; it was right after that.

Okay, so, so yeah, we had—we it was not obvious that we could sell, and we were almost out of money, and I think we got down to, like, 60 days of runway or something like that, and we were already started working on Plan B, and we’d been trying to fundraise for three or four months, and you know, we had maybe we had 80,000 signups at that point, and you know, some amount of usage of that, maybe three of single-digit thousands of teachers using this, so it was interesting but not obviously a great business.

And we talked to, you know, our early investors—none of them wanted to re-up, actually! And we got a couple new investors who were passionate about the vision and who, you know, were much more interested in the change that we were trying to bring about in the world than the us being able to make a really clear economic case for it. But they were optimistic, and so we raised $2 million then, and after that, we hired this VP of Sales and we’re like, alright! So now, you know, we don’t want to ever run into the situation to get... let’s try to start trying to solve the sales problem sooner than we otherwise would have because we still felt like we had small numbers.

So just a quick pivot in the short amount of time we have left: You talked about making something that the teachers love, and at Y Combinator, sort of our motto is making something people want, which, by the way, if you're gonna default to anything as a secret for success, that’s it! And it’s great to point out that you guys are Y Combinator summer 2017 batch.

Tell us about your thought process as a company that was actually doing, frankly, really well when you guys decided to come and do YC.

Yeah, ‘cause you just heard the bad part where you heard 60 days of runway, but when we decided to join YC, we were already profitable!

And it wasn’t—it was the easy decision.

Provide, see? Of 8 and we’d applied before.

We had applied before, I don’t know if you know the hard way.

Yeah, that’s true; we got rejected early on back with the tech that the Pinterest website was—yeah, I don’t know that our story is that telling. I mean, it was a very unique decision for us; it was a hard decision.

There were a number of things that we were—a number of—we were fundamentally trying to sort of get the business to the next level. We thought YC could help in a number of ways. I don’t know. I don’t have a good sense. It was a very much a pro/con list approach of, like, well, we could get this; we could get this, but here’s the downside. I would say it summed up as we were pretty sure we were gonna raise another round about then, and YC could help with that.

We were focused a lot on scaling the team, and we were going—we were planning to do a sort of PR push about that time anyway. Being part of YC would be an interesting part of that. I had heard great things about the program from multiple friends who’d gone through it, one of whom had just finished it and he was a really unconventional company in the batch—the supersonic company.

And so I had a number of conversations with him about that.

Yeah, Blake, because we were unusual as well and so we compared notes on a number of things, and then a supersonic plane of ed-tech company like it and I had lunch with you and you were part of convincing me, at least, that oh, no, it’s not that weird for a company at your stage; like you should consider it.

Anyway, it wasn’t something I was seriously considering at that time.

What is there any particular piece of advice you'd give to folks here who many are all of which are hopeful to do YC as they think about going through the process?

Do it!

Yeah!

I mean, we’re—that advice, well—everyone knows YC is about—you know, the reputation is this early stage, of course, and it’s helping people find product-market fit, and so definitely I like we saw that and it was funny, like, for Keith, who’s a seasoned entrepreneur, I mean, we really enjoyed the talks as we did it, but for me, having never—this is my first time being an entrepreneur—like the talks were amazing! I mean, that alone, the mentors, the network, the advice that we get exceeded all of my expectations.

And so it sort of lived up to the reputation, and I think at the end of the day, it’s a combination of the advice, the clarity on our own business and refining of strategy, all the other founders that we met and Bookface, which is the internal social network, and then the ultimate sort of fundraising process—we raised money after on demo day too.

Awesome! I’m gonna ask one final question for Doug. Doug, you pulled yourself out of a nice, comfortable, cushy superstar science teacher job into the role of a full-time founder. Can you talk about your thought process in doing so and what it's been like?

Yeah, so I think it was interesting earlier to see how much—the, how many people in the room are already doing startups. I don’t think I have to, like, I don’t think there’s any teachers out there where I have to know, yeah, okay, if there are, if you're on the edge, talk to me because like I can help push you!

I mean, there’s sort of advice I can give to teachers about if you’re thinking about doing this. I mean, for me, it was—I was always interested in scale. Now in hindsight, it’s so naive and foolish to think that, I was teaching 60 kids a year, and I was at this private school that had—was going to scale up.

And I remember now, I play it back in my head, and it’s like if we work really hard in 10 years we'll reach a hundred thousand kids! And, you know, if you’re working in tech, like, you should rightly be like, yeah, like, you could reach a lot more than that!

So I was—it was around this time that it was the rise of these science communicators on YouTube, and I remember I always had this sort of mystery approach to my science where we start with things with a question, and this guy Vsauce put out a video in 2011 called Why Do You Have Two Nostrils? And it was amazing! I was about—it was about your respiratory system. It’s totally the kind of thing I would have made, and by the end of the week, he had 1.3 million views, and I said to Keith, that’s the same number of people that watch ABC Monday Night Football!

Like, I just—that many people want to know why you have two nostrils! And like he’s just run into this like accidentally! Like I want to actually—I want to do this!

Like, none of those YouTube Youtubers are trying to make a science curriculum or actually influence science education in the country, so yeah! I was a pretty easy sell! Like once Keith said take what you're doing, bring it up here; the digital revolution is happening. Let’s reach a huge audience.

Thank you guys so much for coming back to spending time!

[Applause]