In the education technology business, Larry Berger is considered—if not the smartest guy in the room, then certainly one of the wiser ones. With more than 20 years in the industry, Larry has seen the ups and downs, twists and turns.
In 2000 he co-founded Wireless Generation, which pioneered the use of data, digital diagnostics and assessments to support students. It was bought in 2010 by News Corporation, which invested more than $1 billion into the company and rebranded it as Amplify. News Corp’s commitment proved to be a short-lived, however. The media giant sold Amplify to private investors five years later.
Today, Larry Berger leads Amplify as its chief executive. The company is no longer as high-profile—or as big—as it once was. So what is Amplify today? What have the past years taught him, and where is the company going?
EdSurge recently sat down with Berger for an update on what Amplify’s up to, along with his thoughts on how the curriculum business is evolving. He also talked about the challenges facing edtech companies today, including his skepticism towards what he calls an “engineering” model of personalized learning.
Subscribe to the EdSurge On Air podcast on your favorite podcast app (like iTunes or Stitcher). Or read highlights from the conversation (which have been edited and condensed for clarity).
EdSurge: What is Amplify in 2018?
Berger: There’s really two things that anchor Amplify today. There’s a body of work we’ve been doing for a long time around diagnosing reading and math issues in young kids—these are observational assessments. Teachers do them on a mobile device or a laptop, or they do them one-on-one with kids. And then capturing that data and being able to give feedback to teachers, to parents, to the system as a whole about how reading progress is happening. And that’s the thing we started doing back in 2003, which has grown and evolved.
Then, over the last decade we’ve started to do more and more in curriculum. And I would say Amplify is increasingly a curriculum company. So we do K-8 English language arts curriculum and K-8 science curriculum.
These are blended programs, so they’re not just a textbook. They’re a lot of interactives and simulations. There’s also a lot of print materials and, in the case of science, a lot of hands-on science experiments that kids are doing, too.
Not too long ago, the question was: “What wasn’t Amplify?” We were talking about tablets, games, books and other kinds of services. As you removed some of that and condensed your focus, what do you think was the biggest lesson that you’ve learned?
Education requires a lot of patience, and the products that become great in education rarely do so because someone [builds] something in a garage, puts it out there and it becomes huge overnight. The thing that tends to happen is you invent something, you try it in schools, and you realize you didn’t quite get it right. Teachers teach you how to make it better, and over a few years you get to something that’s pretty good. And over a few more years you get something that’s a real breakthrough.
I love that journey. I wouldn’t want it to be any other way. I like the learning process, the back and forth. It makes us better.
But if you’re a big company—and we got acquired into a big company—you want to be able to just say: “Can’t we just spend more money and skip all that and build something that’s going to make an immediate impact?”
Before we were part of [News Corporation], we would build things carefully and methodically, and make sure that we really researched things before we scaled them. And I think right now we’re back to that.
As a part of News Corp, there were bankers calling about your results every quarter. Does that time pressure create an unrealistic time-frame for the things that you wanted to build to bear fruit?
Yeah. I think education products grow methodically over time. And then schools take their time piloting, expanding—eventually going district or statewide. But those things are multi-year things that have as much to do with the relationship you establish with the people in the school and the goals that you’re helping them achieve, as they do any kind of marketing or business or publicity.
So I think increasingly we’ve been sort of keeping our heads down trying to build great stuff and trying to respond to school customers.
For a company named Amplify, it’s been a little quiet over the past few years...
I think it was an intentional effort to really focus on what we are doing and let teachers and students start to be the voice of whether we’re up to is great. And the really fun thing is that while we’ve been quiet, with not as much publicity as back in the day, the teachers in the schools are starting to really speak up about how this stuff is helping them.
The big publishers used to own the markets and channels. But how is the K-12 curriculum market shaking up? Is it easier for a company like yours to get your foot in the door?
I would say that there is a certain fatigue with those big publishers, and there’s a sense that we don’t want to buy the same thing that we bought last time. Historically, there hasn’t actually been a lot of choice. The investment that was required has meant that only a few companies with the resources to do that can afford to build that product. And only a few companies could afford enough salespeople out there in enough places where they can meet with the schools in their territory.
When we were a little startup—Wireless Generation—we just didn’t play in curriculum at all because it was too expensive to enter that space. And so those guys get to be unchallenged. But part of what happened when we were part of News Corp was that people said: “Let’s make the investment to create an alternative, to create an upstart in that sector.”
What gives you optimism that you can go toe to toe with the big publishers?
In some states, in some districts, they pilot really extensively. They’re trying the different things and seeing which one’s actually working well for the kids.
There’s a pretty rigorous process of how [core curriculum materials] gets chosen. The challenge is it’s expensive to participate. They want you to send samples, they want you to support pilots. But I think it is actually a pretty meritocratic decision that happens at the end.
Do schools want digital?
We built at first something that was pretty heavily digital. And the original model was that every kid needs to have a device for this to work really well.
What we quickly learned was schools were happier with something that was much more blended, meaning if I want to teach with technology, I can do that. When I want to do something in print and when I want to do a little bit of both, all of that is supported. It’s a much more flexible idea of the technology-to-student ratio.
You and a colleague wrote a paper about 11 years ago that people still refer to as required mandatory reading for entrepreneur, called “K-12 Entrepreneurship: Slow Entry, Distant Exit.” If you had to write an update to that paper, would you still give it that same title?
What I would say is there are a bunch of things that have changed since we wrote it, and there are probably more things on the list that seem to me to be more similar than different.
The big change would be that, at the time we wrote it, there were a small handful of people who were willing to make commercial investments in K-12 companies. “Edtech” certainly wasn’t a thing. It seemed a little too risky at the time and not what schools were doing.
There’s an evolution of what schools are doing, and also the sense of an informed market[place], so people now know that there are edtech companies and that some of them are doing well, and that there are investors that are interested in that. Just telling these stories has been a helpful part. And, there’ve been some successful exits to companies that wake people up to the fact that there’s a business opportunity here.
What’s the most frustrating things to you that hasn’t changed since 2010?
There isn’t a “return on investment” mindset in education—in terms of what education returns [a school or district expects to get]. The fact that hasn’t really changed is the conversation around outcomes and efficacy. There’s probably more talk about measurement, but the idea that as we buy something, we [should] evaluate or even pay for it in terms of how helpful is it in terms of getting results.
I’ve gone around the country being willing to do deals with school districts saying, “Don’t pay me unless I show real results.” And schools are like, “I don’t really know how to do that. Either I buy your product or I don’t buy your product. I don’t have a way of holding you to a performance contract.” I think if we could get that in place, lots of other things would change.
Recently, another piece you’ve written has attracted a lot of attention around the limitations of personalized learning. Have we hyped it up too much and set unrealistic expectations?
There’s a kind of hype, and maybe a misdirection of energy, around what parts of personalizing the learning experience are likely to be the most productive.
In that piece I look at the “engineering model” of personalization. Essentially it says this: When you look at a classroom, it is a not a very well-designed network because there are 30 different nodes, [each of them a kid] who is each processing in their own ways. And there’s only one transmitter: the teacher. There’s not a parallel-processing machine that can give each of those kids exactly what they need at all times.
So why not fix it and feed the right thing to each kid at each moment? Wouldn’t personalized learning be using computers so that every kid gets what they want at exactly the moment, and it feels like one-on-one tutoring?
But we just don’t know enough about how to do that. We don’t know enough about how to measure exactly where kids are in order to know what to give them next. We don’t actually have a great library of the next thing to give them. And it’s not clear that if we did, kids want to be in a mechanistic system, or the teachers want every kid to be on their own little personalized learning journey.
There are a bunch of ways that teachers have always personalized learning in the classroom, scaffolding different kids’ learning so that they get different kinds of support. They have always personalized feedback, and good teachers do a phenomenal job of that. And so shouldn’t the technology be helping with a few more of those things that classroom teachers are trying to do?
The engineering model is asking personalized learning to introduce a kind of foreign object into the classroom, and the interaction between teachers and students can be a bit diminished by that vision of personalization.
