“Education is a big problem,” says Knewton CEO Jose Ferreira. He’s spent his whole life in the education space, starting at Kaplan in 1991. He tried to bring adaptive learning to Kaplan in 1993, but it was hard to do so in a brick-and-mortar situation, an environment in which few people even had computers. Ferreira was ahead of his time, and once technology caught up, he founded the adaptive learning platform Knewton in 2008.
Education might not be the sexiest of industries, but it’s a $7 trillion global industry that’s ripe for innovation. To date, Knewton has received $54 million in funding to build its Adaptive Learning Platform that helps you learn better by responding to what you do and don’t know, personalizing the content that’s served to you based on data collected about your learning habits. In short, Knewton enables smarter learning. Consequently, students are better prepared for school, which makes class time more fruitful.
Knewton has had students in 190 countries, and this fall over 600,000 students will use the platform. Mashable spoke with Ferreira about big data, what inspired his innovative approach to ed tech and the impact he hopes Knewton will have on the world of education and the world at large.
You founded Knewton in 2008 — how did it come about? What was the impetus for starting an adaptive learning company?
I spent most of my career in education and technology, I worked at Kaplan, and I was one of the first people trying to bring innovation into for-profit education. I guess it was ‘right place, right time,’ and it was clear to me what was going to become available, in terms of functionality, and what was going to be possible. Education has always produced an incredible amount of data, that’s always been obvious to me. But technology had to catch up.
The things that I tried to capture with early products at Kaplan — giving people lots of tests in a room with a proctor in a bricks-and-mortar situation that’s difficult and irritating to administer — that problem goes away if people are going to read their textbook on an iPad. And then it’s really easy to capture the data, as long as someone has gone through the work to wire that content properly.
So at Knewton, we convince publishers to wire their content properly, and then we can unlock these vast data sets. These data sets are way bigger than any other vertical market’s data set. Google gets 10 data points per search — we get five orders of magnitude more data per user than Google does. We get more data about people than any other data company gets about people, about anything — and it’s not even close. We’re looking at what you know, what you don’t know, how you learn best. The big difference between us and other big data companies is that we’re not ever marketing your data to a third party for any reason. [The data is collected] to make your education better.
Our personal algorithm is constantly evolving and dynamically generating recommendations in real time. Within a few hours, you get a unique bundle of material that’s targeted to you. Within a few weeks, we know tons about what you know. We know as a percentile how strong you are in every concept in your course. We start to be able to guess really accurately what strengths you have in the concepts we haven’t even shown you yet, based on correlation analysis with the concepts we have shown you.
Learning pathways vary for each student
We can predict failure in advance, which means we can pre-remediate it in advance. We can say, “Oh, she’ll struggle with this, let’s go find the concept from last year’s materials that will help her not struggle with it.” So you’re not just getting a textbook, you’re getting a subscription to this huge content portfolio, and we just go find the perfect little bit to help you master a concept. It can be a video clip, text or a video clip with text if we know that you learn better that way. If you learn quantitative things better in the morning and abstract things better in the evening, we’re going to know that. If you learn English literature concepts best in 25 minute bursts, [that’s what we’ll do].
We know everything about what you know and how you learn best because we get so much data. And education is the highest stakes media product in your life. It’s infinitely more important than your Facebook friends’ status updates or your Google search results, because it’s your future.
What is the end goal for Knewton?
There are a couple goals. It’s pretty clear to us that there’s going to be one dominant data platform for education, the way there’s one dominant data platform for search, social media, etailing. But in education, it’s going to be even more winner-take-all; there will be a number of companies that make up the platform, like Wintel. People might make a perverse choice to use Bing for search because they don’t like Google. But no one’s going to make the choice to subject their kid to the second-best adaptive learning platform, if that means there’s a 23% structural disadvantage.
The data platform industries tend to have a winner-take-all dynamic. You take that and multiply it by a very, very high-stakes product and you get an even more winner-take-all dynamic. So someone has to build that platform. Right now we’re the only ones trying, and it’s kind of amazing to me that no one else is trying. I think part of the reason no one tries is because it’s so unbelievably ambitious that nobody thought it was possible, including our own investors — our investors didn’t really believe in the platform vision until we proved that it would work. They thought I was this crazy entrepreneur, they were very skeptical.
But that’s not the vision, that’s the means to the end. The end is a really simple mission. Only 22% of the world finishes high school, and only 55% finish sixth grade. Those are just appalling numbers. As a species, we’re wasting almost four-fifths of the talent we produce. What if the person who invents a grand unified theory of physics is growing up in Africa and never gets a chance? What if the girl who invents a cure for ovarian cancer is growing up in some Cambodian fishing village and never gets a chance? I want to solve the access problem for the human race once and for all. We’ve always had this problem, and no one talks about it because we’ve always had it. Of the 1.25 billion kids in the world, a billion won’t finish high school. That’s a tragedy, a preventable tragedy.
How does Knewton work in the classroom? How is it used day to day?
We provide tools for teachers to help their students better prepare for class. It’s an ancillary product. Instead of teachers teaching out of a textbook, we give students and teachers access to a huge library of content, and the kids come to class better prepared because they all learned it in the best way for them.
Knewton can power anything where there’s a “right” answer — that’s all of math for much of K-14. We also can power most of verbal skills, since most people agree on what constitutes good writing, like grammar, sentence structure and topic sentences.
The things we can’t power are things like art or ballet or philosophy, where there’s no right answer ever. But even what people think of as pretty subjective courses, there’s often a right answer. In a history course, we’ll take you 85% of the way there, [until you get to] causes — was the Civil War about slavery or states’ rights? The point is to get the students ready for class so you can divide them up and have them argue with each other over the causes of the Civil War. Let’s have that discussion in class. Let’s not waste time on all of the facts and dates and concepts — Knewton can teach those better.
So as a teacher, you get tools you never had before. You can tell students how many concepts they mastered and what they should work on over the weekend. And on day one [of a new school year], we can tell you what everyone’s good at, what they know, how they compare to the rest of the country on a percentile basis, whether they’re a visual class, and we can offer insights for parents. Teachers will be able to add their own content, upload it, tag it and seamlessly use it. And they will be able to search the platform; Knewton will psychometrically determine which are the most effective pieces of teaching coordinate geometry.
Knewton uses gamification for learning, offering badges and points. How does that help students learn better?
There’s nothing magical about games. Games are slower than text, video is slower than text — text is the fastest way to learn. Games are a backup if they don’t get it right the first time. Let’s say we give you a couple geography questions, you get them wrong, then we give you a game because we think you might do well with a game. Exact same concepts, just taught in a different way, and now you get them right. Did you get it right because it’s a game, and you learn better in that way, or because you saw it twice? No one knows.
Knewton knows every concept that’s highly correlated to those gepgraphy concepts, so when you get to those correlated concepts, we’ll give you the game first. We can algorithmically ask ourselves and answer why you got it right, whether it was because you saw it twice or because it was a game.
For Knewton to work really well, we need lots of content, and we need different types of content. So we do massive deals with publishers. Our goal is to open up the platform in the next couple of months to let any teacher in the world find and upload content, and Knewton will recommend the most effective kinds. Maybe there’s a kid who’s not good at math, and if he sees it in a game, he gets the confidence to keep going.
We can also trick you into working harder with a game-like interface. We have points and badges, which we use as “microrewards.” Adaptive learning will optimize every minute you give us. Adaptive engagement is a way to trick you into giving us more of your time to hit more material and learn more concepts, because we can measure your engagement. We’re going to constantly tease with “Oh, you’re so close, keep going, great job,” to keep the student engaged.
How has the education technology space changed since you started Knewton?
When I started raising money for Knewton, there was virtually no interest in ed tech companies, and no one was using the term “adaptive learning” that I could find. Now it’s sort of this buzzword. 2tor and I started at the same time, we talked to the same investors, and I think we were the only people raising money at the time. But there has been a little boom in ed tech. I go to conferences now that have hundreds of companies and they used to have just dozens of companies. The industry is changing really fast.