Know think do

The way I think about learning is in three parts. First, you have to know a bunch of things to be successful in life. There is no way to fake it if you dont know multi-digit addition. At some point, if you don't, you will put the wrong things into your calculator and give $20 change on an $18 purchase. (This one happened to me last year). So there is a good body of things you have to know. I think this maps cleanly in k-5 where its pretty inarguable that you need to know how to spell and most everything else they teach you. On the other hand, in secondary grades, it is not as clear. Is it crucial for everyone to know biology? Maybe the amount you can look up on Wikipedia, but a whole year? We can debate what things you need to know (which we do periodically, common core being the latest output), but its clear there are things you have to know.

Second, you have to be able to think. You have to use the things you know to solve problems, sometimes in groups, to create new things, to identify patterns. You can be taught how to think. If you aren't sure of this, talk to people who went to engineering school or law school. The methods great schools teach you about how to think are far more valuable (and more difficult to assess) than what they teach you to know.

Finally, to be a successful person, you have to be able to do things. Getting things done has everything to do with your personal and social abilities. Are you persistent? Are you organized? Can you delay gratification and focus on the long term? Do you work well with others? This third category is much harder to measure than the first two but probably accounts for well over 50% of your success in life, especially modern life. Read Reid Hoffman's - the startup of you - sometime to see the product of a very evolved doers mind. Your career and your life is less shaped by the specific things you know (beyond what you learn k-5) and more by how well you adapt to a rapidly changing world. So knowing less and being able to think and do more is going to serve you well.

In this context, we are getting ready for online learning 2.0. Version 1.0 was the simple translation of teaching and learning online by the pioneers like khan, coursera, udacity and others. Pause and rewind is not what online learning 2.0 will look like. What will it look like? Obviously this is what we are building at my new company, so I'm not going to pre-announce anything specific, but I do think there some obvious things. Online learning will teach you almost everything you need to know. Some of that will be through hundreds of hours of practice. Other will be on-demand (arguably for a lot of what you learn in secondary). But anything that requires personalized attention and infinite patience on the other end of the student will ultimately go online. Physical schools, when they are available to a student, will teach you how to do. Physical places and groups will have big advantages in getting students to change their behavior in the way that these personal and social skills require. In the middle I think we will see online learning systems get better and better at teaching you how to think, but this will likely be the domain of institutional learning for a while.

Of course, it is not quite that simple, but its important to acknowledge the relative strengths of teachers and tech so we can think about how they work together rather than the current debate about how tech will replace teachers. Teachers aren't going anywhere, but their roles will change dramatically over the next two decades. They will be much better paid (because the role will be differentiated the same way professors and TAs are differentiated), they will spend much less time teaching rote skills (the knowing, because that will happen online) and much more time teaching children how to think and how to do. I think we will end up with much happier teachers because of this change, but of course change is hard.

So to me this frame of know, think, and do helps you get clearer on how learning will change over the next 20 years. Now that everyone is online (or coming online faster than you can possibly grow your company), online learning companies have enormous market opportunities. I think we will find that the quality of the learning products they build for the 8 billion people out there makes what we have now in edtech feel like the comparison between the VCR and Netflix.

Universities - Get Your Ed Schools Involved Quickly!!!

One of the most striking things to me about the higher education side of online learning is how poor these classes are.  It's really not that incredible that a tiny percentage of students ever finish a class.  Now at this point, university people are saying, "Hey, this is no worse than our classes have ever been!"  Exactly!

The big change here is that universities previously were built on the research reputations of their faculty.  The fact that they actually had to teach was a burden.  Believe me, my wife was a professor for many years.  She actually loved teaching, but she was far from the norm.  Think back to your college experience, how many professors would you give an "A" to?  It was probably strikingly small.  Going through engineering school, I would give an "A" to many more of my TA's than the professors and many more of my peers than my TA's.  So what the universities have put online is literally their worst work - the work of professors who often know nothing about teaching.

I was on a panel a few weeks ago with a couple of MOOCs.  I got tired of hearing about how millions of students were taking these classes.  It's nice to have this much low hanging fruit, but your product can still suck.   So I finally asked my other panelists what they did about students in their classes that were struggling.  The most honest one actually said, "That's the student's problem, our job is to make sure we explain the content clearly."  This insight into what higher ed thinks teaching and learning encompasses is incredibly illustrative.  For the other 99% of students (not the ones the elite universities are used to "teaching"), every student in your class will have large gaps in their knowledge that keeps them from understanding what you are presenting, no matter how clearly you present.  They won't be able to scurry around and fill all of those gaps quickly enough on their own to finish your course.  Filling those gaps is actually what you do for a living.

I strongly encourage everyone in the higher ed space who has a broad mission to educate the students of the world (and not to just educate the ones they already educate, but do it online) to go out into the real world and figure out how to teach the other 99%.  Udacity's work with San Jose State will be the greatest asset they develop for that company if they make it out alive.  It is extremely likely that if they really get it, they will end up completely redesigning the process of teaching and learning in their system.  I really hope others and the universities themselves do the same.

Here is the ironic part of the whole situation.  The education schools at universities, often the ugly stepchild of higher ed, actually have people who spend their time understanding how the other 99% need to learn.  It would be really great for people to walk over to the ed department and ask for help.  I'm not saying ed departments are founts of knowledge, but they at least understand that the type of teaching happening for kids in k-12 gets much closer to what it takes for students to actually learn.

Who's your customer?

One of the things I have been struck by the past few edtech gatherings is that the traditional education technology industry doesn't seem to know a lot of teachers or students.  Coming from the world of teaching, this struck me as pretty bizzare.

But I realized, because the person with the checkbook is the superintendent and to a lesser extent the school principal, these companies know the superintendent in general and only know other people if that's what it takes to make a sale.  The likelihood that they are going to get to know teachers or students well is unlikely (other than their support team on occasion).  That's pretty broken!

It seems to me that one of the only ways we are going to get software into classrooms that actually does things which are useful to students and teachers is if we change the customer.  I don't want to be too simplistic here, but why don't we make the student (and their parent) the customer and the teacher our partner (i.e. the one that let's us acquire customers).  Now the only problem with that is that it completely screws up the revenue streams of the existing companies, because superintendents aren't paying, school leaders aren't paying, and teachers aren't paying.  But really, other than the people running those companies, who cares?  Once students are your customers and teachers are your partners, you are going to build things they like.

Just to make one last argument to the people actually trying to change the outcomes for kids through the software they build (and I'm going to be optimistic here and say that is why most people in this industry build software for education rather than just to make money).  Let's look at this industry for a moment.  It's about a $4 trillion industry globally.  It spends between 1-2% of that on technology.  Anyone who thinks that number is going to get larger needs to learn more about the politics of public education.  So if we do the math, institutions spend $80B or so a year on technology (hardware, software, and services).  That's not too bad.  Unfortunately Pearson, HMH and News are going to get well over half of that.  So everyone else is battling it out for $40B.  Compare that to making students your customer (i.e. focusing on the consumer side of the learning market).  Just to start, tutoring and test prep is $100-$150B per year globally, children's books, apps, etc. are all large markets.  So the consumer side is probably 4-5x larger today (before the shift to more direct learning by students has really happened) with much less friction than the institutional market.

So I will say it again.  Other than the fact that this is not what we do today, why not make the student your customer?

Digital video editing

A friend of a friend told me a great story about being with avid when they brought digital editing to the film industry. The first time he showed the system, the editors said, "you don't understand, if the director comes in and we aren't cutting film and taping it together, he will think we aren't working". It took him two years to convince them that they were experts in composing scenes and not cutting film. Remind you of any other industry? Except 20 years later?

The Shift from Institutional to Individual Learning

As I get my new company going, I've been reflecting a lot on exactly why online learning has been so bad for so long.  I keep coming back to the fact that if I had to start this company today, and sell our online learning system to institutions - schools and universities - it would be a slow and arduous process.  Since the venture community (speaking mostly of traditional high-growth silicon valley venture) understands that education institutions are bad customers, capital would be scarce.  Since the institutions are hard to sell to, I would have to field a pretty significant sales force.  The combination of lack of capital and high sales and market expense just starves the ability to put a lot of money into building great products.

So what has changed?  People are online.  Since curiosity is a basic human condition, it seems pretty clear that when people get online, they are going to want to learn online.  They aren't going to have to trudge down to some building somewhere to find the sage who can teach them.  They are just going to start learning online.  Learning is a ridiculously large business globally, accounting for about $4 trillion of spending each year.  To date, most of that has been going to the institutions, because they were the guardians and providers of knowledge.  In an industry this large, with 2 billion students under 18 and arguably the other 6 billion folks in the world also curious, it doesn't take a very big shift in the way we learn to open up massive opportunities.

Like everything else with the Internet, it seems very likely that the shift in learning online happens long before the spending shift happens.  I think arguably we are already seeing a huge amount of time shift happening, with people learning through Khan, Udacity, etc.  Every time that kind of shift in time has happened from traditional industries to the Internet before, the spending shift happens later.

But here is what is exciting.  That market problem that kept innovation from happening in online learning is going to decrease year by year as learning moves to the individual, and the amount of innovation is going to increase year by year, until individual learning is much more compelling for a variety of things than institutional.  That rebalancing of institutional and individual learning to more accurately reflect their strengths - institution (social, network, collaboration) and individual (personalized, self-directed, student-centered) - is going to create enormous dividends in the number of people globally who get a solid education and the number of people who can go further in their learning sooner because no one is telling them it's age inappropriate.

Kids learn when they are solving problems

It's funny, sometimes I have the same conversation 10 times in a couple of weeks.  Lately the conversation has been about the nature of how kids are going to learn online.  I think the idea that kids are going to learn from pre-recorded video of a teacher discussing a topic is silly.  I have never seen a child at Rocketship learn that way and my friends running secondary schools say the same.  Maybe by late high school or college this changes, but I kind of doubt it.  When you are in a school, I think it becomes very clear when learning happens.  Students who are working on a problem that they can solve learn by trying to solve the problem and receiving prompts and insights from peers or the teacher when they make mistakes.  This eventually helps them get over the hump and be able to solve similar problems with a lot less mental effort.  That's learning.  This happens thousands of times a day in well run classrooms.  For whatever reason, we have really lost this truth in online learning.  First, we don't spend near enough time trying to make sure the student is trying to solve a problem that they have the ability to solve.  Second, when they make a mistake, most of the time a buzzer just goes off and we hit them with a similar problem to frustrate them more.  We lose the teachable moment all of the time.  And finally, when you watch the amount of receptive learning students are expected to do outside of the context of problem solving to "learn" what to do, it often dwarfs the active learning time.  I am really hopeful that the next generation of online learning systems approaches learning from the perspective of the child solving a problem they can solve.

Why Blended Charter Schools are "Whales" in the ed institutional context...

My past few posts, I've painted a pretty grim picture of what it takes to sell to schools and districts.  I want to take a moment just to answer the question "But hey, Rocketship doesn't act that way, you work with a ton of companies, you pay for products, etc."  It's actually taken us a surprisingly long time to figure out just how different we are.

The fundamental reason we are different is that we are completely comfortable with the idea that we will use the best method for every type of learning for every student.  What this nets out to for edtech is that if there is a program on which students learn certain concepts faster than with a tutor or from a teacher, we are overjoyed for them to learn that online.  The efficiency of online learning (i.e. its cheaper per minute of learning time) allows us to redirect resources to pay teachers more, support them better, and put even more effort into the code we write to understand our students better.  Because we are incredibly incented to allow students to learn as efficiently and effectively as possible, every time we can use edtech, the cost of that program is insignificant on a learning per minute basis compared to our other options.

So while traditional schools often having no budget line for edtech and having to steal money from their curricula line item, any serious blended school has a budget line for edtech.  Ours is $100/student.  Before your mouth starts watering too much, that $100 gets cut up among dozens of programs for assessment, instruction, scheduling, etc.  So it's not a goldmine, but the point is that we have a line item.  That means when you come to us with something good and we test it and it works, we want to buy it just as much as you want to sell it.

Compare this behavior with traditional schools.  Traditional schools are not agnostic about learning methods.  In general, they believe that a teacher with their class of students is the one and only best learning approach.  If they happen to have computers in that classroom, the number of kids learning on those computers has nothing to do with the number of students that teacher can impact.  So economically at least, they are disincented from purchasing technology.  The school has already sunk the cost on its teacher salaries, and providing better online learning might be nice for the teachers, but since the perspective is that learning comes only by teachers teaching, the technology is just an afterthought.

So that's basically how blended schools are different.  For many years, we implored vendors to build things for us and we convinced quite a few.  What we've realized is that we aren't a very big market yet.  Even within the charter sector, blended schools have only started to take off, there are probably fewer than 100 truly blended charters at this point.  And on the district side, I have not seen a district make labor trade-offs yet, meaning that technology is not strategic, just a support for the teacher doing everything themselves.  So a market of 100 schools with maybe 50,000 students is not nearly big enough to build a company around.  How quickly does this change?  My guess is that within the charter sector, this begins to change extremely rapidly (for our world), but nowhere near fast enough to worry about for the next decade.  I would not be surprised to see 1000 blended schools in a decade, representing 500,000 students.  If you had a great product and could charge $5/student, that would be $2.5M in revenue.  Not so good.  Outside of the charter market, is there a chance that things flip over quickly to blended and we see edtech become strategic?  I'm usually pretty optimistic that change can happen, but unfortunately this goes right at one of the sacred cows in education - that only a teacher can provide the learning a child needs.  There are many forces between unions and districts and school boards that will be very reluctant to rethink this belief. Even when Rocketship and other blended schools are getting 2X the learning for kids and 2X the salaries and job satisfaction for teachers, I think those institutional changes at traditional schools in district will be painful and slow.

Higher ed is a really good analogy here.  The new wave of online higher ed companies like Udacity are going directly to students (for free!).  While there may be some universities that get it and will be able to address the needs of students as well with a blended model (EdX from MIT and Harvard for example), the vast majority will not make the transition and there will be a new way that students learn after high school.  This undoubtedly will make access to quality courses much higher and create a lot more equity than anything we could do physically (absolutely everyone is going to have a smartphone before the traditional institutions get on board).

So as a K12 edtech entrepreneur, I think that if you make the student your customer, you are in good shape.  Schools like Rocketship will be great ways to validate and test products and we are happy to do it, but ultimately, you are going to make your money by doing a great job with the students themselves and we are just a channel to the student.