Interview with Keith Schacht

Co-Founder, Mystery Science

Keith has been building start-ups since college, and a common thread among most of these ventures is they involved teaching people about technology or science. He took a break from entrepreneurship and joined Facebook as a Product Manager, where he worked on News Feed and Messenger. In 2013 he left to launch Mystery Science.

Mystery Science builds science lessons for K-5 teachers. Their goal is to make it easy for teachers, many of whom don’t have science backgrounds, to deliver inspiring science classes. Modules are centered around a hands-on activity that asks kids to solve a mystery of particular scientific phenomenon. They provide teachers with explanatory videos, supplies lists, and straightforward instructions — all the pieces to needed to create a great lesson without extensive prep.

They believe scientific curiosity is innate in most young kids, but it gets muted in grade school. By framing science as a series of mysteries, teachers can give their students a method of exploring the world, not just a series of facts. As those kids discover new phenomenon, they now have tools to evaluate information and come to their own conclusions.

—

Let’s start with some background. What kind of school did you go to? What kind of education did you have for the first part of your life?

I was a military brat, so I moved around and changed schools every few years. I learned a lot on my own and most of the things that I’m passionate about today come from side projects. My biggest takeaway from K-12 was that I didn’t particularly like school. It felt like an interruption from projects I was working on at home.

I liked to build stuff. My first projects involved cardboard and pipe cleaners, and I soon graduated to wood and saws. But things often fell apart and that was frustrating. You’d build a go-cart but the wheels would come off. Sometimes it would work, other times it wouldn’t. When I discovered computers, part of the joy was that it didn’t fall apart, and it did the same thing every time. There was an unpredictability in physical contraptions that got much better with technology.

Was there an experience in school or a teacher who was particularly memorable to you growing up?

I was fortunate. A few schools that I went to had gifted programs, and I got to work more independently. In middle school, I convinced my school to buy an IBM PC so I could learn to program C++. I went on to write BBS software and run a BBS out of my bedroom.

I also have a strong negative memory about school. I had an AP History class, and it was very hard. I put in incredible effort, memorizing everything; dates and timelines, who fought what battle, and who won. I didn’t do very well and the process felt so different than the engagement I got from my projects. Today, I remember nothing from that class other than that feeling.

Did your feelings about learning change when you went to college? Did you become more engaged?

I went to the University of Illinois and focused on computers, interface design, human factors, things like that. I was excited about college, because more of my school time was spent on stuff I was interested in.

I began building web pages on the side. It was early in the web and people paid well for that kind of programming. I had a server under my desk and hosted websites. It quickly went from a part-time job to a web consulting business. So I brought in a co-founder, and we hired other students to help us. By my senior year, it was monopolizing my time and I dropped out of school.

I fell in love with the experience of building a company. It was sort of the ultimate project. Web consulting also involved explaining technology to people. That kind of teaching was the most enjoyable part of the business, and so I decided to focus on that next.

I sold the company and co-founded Inventables. We would research new materials and technologies and then teach R&D groups how to apply them to products. Our clients were companies like Nike and P&G, and we spent most of time explaining stuff to engineers and product designers.

After a few other ventures in games and blogging, I joined Facebook. I became a Product Manager for News Feed and then Messenger. That was the first time I had worked for someone else, and I got to learn how they made decisions and how this incredible company was managed. It was a great experience.

Did you leave Facebook with a clear idea that you wanted to work in education?

I left Facebook to return to the path of teaching people about science and technology. I had also become a parent and that made me more interested in what was happening in school.

I started Mystery Science with a college friend, Doug Peltz. Whereas I had pursued engineering, his passion had always been the natural sciences. After college, Doug went to the classroom and taught elementary and middle school for eight years.

We knew we wanted to work on science education. When kids are three or four, they are naturally curious about the world. They are fascinated and will ask an endless stream of questions. But by middle school, most kids have lost that interest, that curiosity. Children don’t wake up one day and decide that they don’t like science. Something happens in the elementary years to disrupt their interest.

We didn’t have a specific product figured out, but we did know that it was hard to teach science well. In the lower and middle school grades, we heard from many teachers that it was their most difficult subject.

Why is it so difficult to teach science? What makes it different than other subjects?

If you had to step into a third grade classroom tomorrow and teach, the average person would be comfortable teaching third grade math or reading.

But teaching third grade science is harder. Take something seemingly simple, like the like the Earth’s composition. You might say, “there’s the crust, there’s the mantle, and the molten core.” You can show diagrams and give definitions for each of these words, but that’s a vocabulary class, not science. Children should learn that people have only drilled seven miles into the Earth; we haven’t even gotten through the crust. How do we know there is a molten core in the middle of the earth? This is the interesting conversation to be having.

Every scientific conclusion started as a mystery in the world. Someone noticed something, and they were surprised by what they observed. Then they worked to figure out what was going on; why was that happening?

As another example, the sun rises and sets every day. It comes up on one horizon, it moves across the sky, and it goes down on the other side. People observed this for thousands and years, and they wondered, “Why does this happen? Why is the sun moving across the sky? Is there a chariot pulling it across the sky? Is there a sphere that’s turning?” There was a long process of debate and collecting evidence and integrating that evidence to reach a conclusion. You can’t simply jump to the conclusion. Children develop a scientific perspective on the world by understanding how we know the things that we know, not just being able to recite the conclusion that someone else figured out.

It’s like when you’re reading a mystery novel, you don’t just read the last page. Sure, now you know who committed the crime. But without going through the mystery, the set-up and the collecting of evidence, the conclusion isn’t meaningful. It also ruins the mystery.

It’s important to teach science like a set of mysteries rather than a set of conclusions.

Why is that process valuable for kids?

A lot of parents and teachers ask, “What’s the harm in teaching these children these facts? They’re true facts, right? We’re giving children a basic scientific literacy. They’ll know there’s a molten core. They’ll know it’s not the sun that’s moving across the sky.” But when you teach children these facts without teaching them how we know they’re true, you’re doing them a disservice. You’re teaching them that these things are true because your teacher or textbook asserted them as true. That’s the opposite of scientific thinking.

You haven’t equipped children to evaluate new claims they hear. That’s what will help them in any career or subject they choose to pursue as an adult.

If you teach science properly, you give children both the facts and the process. They get a grounding in the facts, which they can then use to reconcile future facts. But by teaching those facts through the method of how those conclusions were formed, you also teach them the process. They are able to reuse that process in the future.

How did you figure what product to build? Was it clear at the outset and how did it evolve?

We talked to many teachers to understand what challenges they faced teaching science and how we might be able to help.

Most elementary teachers said that science was the most difficult subject to teach well. They are usually responsible for every subject — reading, writing, math, and science — and science often caused them the most anxiety. Without a science background, everything is harder: from developing experiments to accurately answering questions from curious students. Good science lessons also needed disproportionate preparation. You need to get supplies for student activities and run through experiments beforehand. The overwhelming message was, “I would love help with teaching science. It requires a ton of time to do well, and I want to do it better.” The hard part was figuring out how we could help without being in the classroom with them.

At one extreme of the spectrum, you can teach the class for them. That might be a science episode like Bill Nye or Mr. Wizard. The teacher can press play and the episode is doing the teaching. At the other end is an outline of a lesson plan. It might include an activity, some points to make, but the teacher still needs to figure out what they’re going say and how they’re going to implement it. Most of the curriculum products fell into one of those two extremes, press play or lightweight lesson plans.

We wanted to play in the middle. We wanted to support teachers more than a lesson plan does, but not so much that teachers gets disengaged from the teaching.

We went through a lot of iterations and prototyping with teachers. How much information to provide and in what format? Should we use video clips? Should we use slides? Should we include descriptive text or just main takeaways? Should we introduce the lesson on a video? Should the teacher introduce the lesson? They need to execute the hands-on activity. We can’t do that remotely, so how do we support them in the hands-on activity? Should it be worksheet instructions or video instructions?

The essential question we were solving was, “How do we support the teacher as much as possible and still be clear that the teacher is the one doing the teaching? How do we help them do what they do best, which is facilitating the learning process and tailoring the lesson to the students in front of them?”

Did you choose to start with a particular age or grade?

Well, our original hunch was to build something for middle school, because that’s where you have dedicated science teachers. But we soon discovered that elementary teachers were more eager for help — they weren’t science teachers, yet they were expected to teach science.

We started with third grade because that’s when science begins to get hard to teach. At the younger ages, science is more about observing and identifying things. Once you get into second or third grade, you’re starting to connect clues together to explain things; the ‘why’ behind various phenomenon. That’s complex to explain and hard to demonstrate well.

In the course of building Mystery Science what have you learned about education that you didn’t understand when you started? Are there assumptions you had that turned out to be incorrect?

I originally thought about K-12 as a single thing because the industry talks about it that way. But that’s not accurate — K-12 as a single market is a fallacy. 2nd graders are so different than 6th graders who are different than 10th graders. Their approach to learning changes as does teaching and the nature of the classroom. If you design tools for teachers or students, you can’t build for K-12 and be really useful. You need to be more focused than that to start.

We also had to learn the business side of selling to schools. We when started, one person described educational sales as “an enterprise sales process with a small business price tag.” Historically, new curriculums were purchased by districts, and those would be selected from a short list of state-approved curriculum. It had the complexity and the timeframe of an enterprise sale, but the economics were modest. I assumed it would be tough.

But that sales process is starting to change. It’s getting easier and easier to go straight to teachers — technology is helping that, and states and districts seem more flexible with teachers finding things on their own. The result is that Mystery Science and other companies are able to take a bottoms-up sales approach that wasn’t possible 5 or 10 years ago. If going through districts was the only way to get new classroom materials into the hands of teachers, that would have been really hard.

You mentioned that becoming a parent made you more interested in the education space. How has that evolved as you’ve worked with more teachers and classrooms?

One of the most interesting questions is the role parents can play. Parents care a ton about their kids and education, but there’s almost a church and state disconnect between the two worlds. We get periodic news about what’s going on in the classroom, but we’re not coordinated or working together. There’s huge potential for parents to supplement what’s happening in the classroom and to continue the discussion, but it requires coordination between elementary teachers and parents. That’s not easy.

When my son comes home from school, the typical questions are “How was your day? What did you learn?” That information is hard to get as a parent, and it’s not another responsibility to simply put on elementary teachers. It’s too much work when there are 30 parents and you’re one person. Companies like Remind are starting to build important communication infrastructure, but it’s still all very early. We bit off one small piece by providing teachers with a pre-written email for each lesson. They can send parents a message saying, “This is what your child learned this week in science, and here’s a great question that you can ask them over dinner to get them talking about the topic.” That’s a very modest start, but parents could be a critical and coordinated part of the education experience in a way they are currently not.

Also, being a parent has challenged how I think about education. Most of us feel passionate about our kids and their learning experience. But when you ask parents to be more specific about the goals or what makes a great education, there’s less clarity than you’d expect given how strongly they feel.

Ask yourself what’s the purpose of education? Are you optimizing for your child’s happiness? Are you optimizing for the college that they get into? Are you optimizing for the number of options they have as an adult? What about their ability to face new challenges? Poll a group of parents who all agree that education is very important, and you’ll find they disagree about what the purpose of education is. That lack of alignment makes it difficult for education to improve because people are pulling it in very different directions.

Having competing goals make it difficult to design better learning experiences. You need some clarity on what you’re optimizing for to see how you’re doing and to make improvements. Without this, the default becomes test scores. That’s the dilemma we find ourselves stuck in. We’re simultaneously very critical of test scores as a measure of education quality, and at a loss for anything better. Until there is stronger alignment on the purpose of education, it going to be hard to make progress.

In addition to Mystery Science you’ve also been involved in opening schools. What have you learned through that experience?

Well, there are two categories of improvements to education that are important to work on.

The first is figuring out the ideal learning experience. If you could control every aspect of the day, what’s the best way to do it? I think that’s really important work to be done because we shouldn’t accept all of the constraints we’ve inherited. LePort Schools and AltSchool are two companies rethinking the classroom experience in this way. These are important experiments to run, and they’re critical to improving education in the long-run.

But I also have a child who is in preschool now and is going to finish schooling before many of these experiments have played out. So it’s just as important to see what dramatic improvements we can make to education quality within some of the current constraints. If you want to change things for 30 to 50 million kids in the US over the next few years, you do it inside that framework. That’s where Mystery Science operates. Most schools and teachers have certain constraints, so we want to dramatically improve science education quickly by working within those constraints.

That’s great. Thank you so much.