The Folly of Scaling the Best Teacher

One of our SchoolHouse teachers said to me recently 

"An educator's truest job is helping them (students) figure out how they best interact with information, the benefit of this system (micro-school) is that it allows us as teachers to get into the dirt with the students to help them figure that out."

The common theory in most educational technology discussions is that we should “scale the best teacher” to as many students as possible, that is how we’ll solve the achievement gaps, find the best teacher in every subject and have every student attend this teacher’s class. Simple right? The internet makes this easy, let’s just put the best teachers online. 

But this point of view mistakes what makes the teacher valuable to the student, and the embedded assumption here is that the way a teacher disseminates information is what makes them valuable. Scaling the best teacher is thinking by analogy and not first principles of what constitutes a great education. 

We’ve done this before, our society gets caught up in ideas and some concept becomes all the rage in a period of time. When our current educational institutions were founded industrialization was popular and working miracles. Look at Henry Ford and his assembly line, look at how many cars it can produce! As industrialization spread, we found all sorts of great products were now available. And so we built a school system that embodied values of the day, and it had success in things like reducing the illiteracy rate, building the middle class, etc., Over time the institutions carried these values forward, but the world changed and now we all question it, “Is grouping by age group the best way to learn?” “Is 30 students in a classroom really the ideal class size?”

Today’s adored idea is scale, driven by tech culture, and so we look to apply the miracles of scale to everything including education. Yet, we’re making the same mistake today with the adoration of scale as we did with industrialization, we’re solutioning instead of looking at the first principles of what makes the teacher valuable to the student. 

What makes the teacher valuable to the student is attention. Direct attention, the kind of attention a great friend or a parent gives. You see, scaling the best teacher has an embedded assumption that the true value from the teacher is in the dissemination of information, i.e. activities akin to lectures, but the true value is in the consultative and adaptive aspects that the teacher provides to the student. It’s in their ability to see the gaps that the student has and derive ways of altering the lessons to make sense to the student. 

This is why we haven’t scaled the best managers, or the best chefs or the best parents or boyfriends. Like these other professions, teaching is partially a consultative act, and it’s quite hard to scale consultation. And it’s potentially impossible to scale human relationships. 

You see the true value in the teacher to the student is in the consultative relationship that occurs over long periods of time to ensure that the student is progressing in the subject matter and growing as a person. 

We’ll talk about one aspect of this consultative relationship between a teacher and student - specifically how the teacher is perfectly set up to address the unique gaps of the student in the next post.

Can a School Have Product-Market Fit?

Scaling a School pt. 1 (~1,500 words)

This “Scaling a School” series is a topic I will return to intermittently in the newsletter. The goal is to expose some of the tensions involved when scaling a school & explore how different institutions resolve them. We’ll start by exploring concepts like product-market fit & the best practices of program design and eventually cover most internal functions.


I’ve worked with many schools and bootcamps over the last decade, and one of the things that has surprised me is that none of them have product-market fit.

To be clear, I don’t mean this to say that all schools are bad or that the system is broken. I’m merely trying to find a provocative way of pointing out that they can’t have product-market fit. Fundamentally they aren’t able to, no school is.

The “strong market demand” from the student isn’t for a school and in the bootcamp space specifically, the “strong market demand” is for a pathway to a better job, and plenty of people wake up needing a better one of those.

This pathway (commonly referred to as “an offering” or “course”) is the connection between the student and the job, and its basic function is to meet the requirements of both. Once it’s able to reliably guide learners to a job it has product-market fit.

But the pathway does all the work here, and it’s success tells you very little about the rest of the school, it simply means that this single pathway works pretty well.

The challenge is that across industries, both the quality of the student population and the requirements for a job vary wildly. The minimum competency level of a Jr developer, for instance, is a much lower bar than say a surgeon, and the incoming students will have very different needs as well. We’re going to spend a few posts examining how changing the input (student) or output (job expectation) requires different structures to support the student along the path.

Essentially what this means for a school is, once it moves past a single program, the school as an entity can no longer have product-market fit: it can only collect pathways— aka “courses”— that do. 

While this distinction might seem pedantic, misattributing PMF to the school leads to misconceptions about the real capabilities of a school and can cause serious issues at scale. 

Let’s dig in.

Misconception 1: A Great School Can Teach Anyone

When you think of “top schools,” what attribute do they all share?

It’s probably not that they have the best teachers or even the best curriculums. Maybe that tuition is too damn high. Even though I didn’t ask anyone, I’d bet “abnormally tight admissions standards” would be on the list. It’s a high bar to get into a top school, and few make the cut. 

And while exclusivity can have many brand benefits, one important pedagogical by-product is that it’s waaaaay easier for a school to have better outputs if they tightly control the inputs. It’s all about the student pathway fit after all.  

For instance, let’s say you were teaching a bread-making course that’s 6 weeks long, and the students expect that they’ll be able to come out of the course and bake some classic loaves from the Great British Bake-off. Which of these students do you think would do best in this class?

A. A pastry chef

B. A carpenter

C. A 6-year old?

My answer is the pastry chef. 

A group made out of pastry chefs would likely be the easiest group to train, followed (probably) by the carpenters. It’s unlikely that the 6-year olds will ever match the first two groups in terms of baking competence, especially in 6 weeks. That would take a very different program design “path”—likely a longer one and perhaps one with a lot of musical numbers and hand puppets. 

The basic principle here again is that some students will have an easier time on the path than others, and so the path has to match the student.

An example of this below:

What can happen for small schools—like a bootcamp—is that they tend to scale by adding more student types to a proven path. Once they have a single successful offering, their natural instinct is to “scale” by putting as many students onto the pathway as they can—whether or not those students are a good fit. The thinking is the institutional “school” has figured this out (PMF) and that in order to grow it just needs to “just stamp out more classes”—fast.

A note from Dr. Michael Hanson

“Also worth noting is that much of education in traditional settings is the influence of peers – if the majority automatically maintain the high standards during the education process that the entrance standards set at the beginning then this feedback loop will have it’s own effects on the pathway as well.”

This is why, on a small scale, bootcamps can be wildly successful. And if they can accurately filter their incoming students for pastry chefs (and learn how to handle a few carpenters in every class) than the quality can hold up over time.

But tension arises with efforts to scale. Even if there are a lot of pastry chefs in the market, they’re not all ready to take your course right now, and so the easiest way to scale is to reduce the focus on student/path fit. Once the admissions organization begins to let in more carpenters, and maybe some 6-year-olds, the next thing you know, finger paint is everywhere.

Most schools don’t recognize this as it’s happening and only notice the after-effects; they think they have a paint removal problem and miss the filtering problem in the admissions department.

The initial impact of poorly-fitting admissions standards is higher dropout rates, more stress on teachers, and typically attention diverts from top students to keep the bottom layer progressing forward on time.

In the future, we’ll cover pedagogical strategies to handle a diverse classroom such as differentiated curriculums and “low bar / high ceiling” but the fact is few schools have the bandwidth to implement such structures now—and no one that I know of has started this work.

If the school continues to admit students who are a bad fit, the good teachers will burn out. It’s not crazy to hear of 50-80% teacher churn in these scenarios, and those instructors who are left tend to water down the curriculum, just so students can pass through without much trouble. At this point, they indeed could indeed teach anyone, but it might not be worth it.

Misconception 2: A Great School Can Teach Any Course

Just like the belief that “our school can work for every student,” this great school fallacy tricks us into believing that we can serve all markets and all topics as well as the flagship offering. It’s all “learning” after all, and we are a great school.

Again, if you assume that the school has PMF, this mentality makes a lot of sense; you’ve got a special sauce that made the first offering go so well. However, this same type of thinking would be like assuming that because Netflix has a hit baking show that every show they produce will be a hit—and a quick perusal of “Nailed It” will hopefully prove my point. 

What typically happens is that more courses are “stamped out” by closely mimicking the original path. It feels safer and easier to copy what works than to reason from the first principles about what a topic and input student requires.

In this haste to duplicate the success, it’s often overlooked that different subjects require different tools, timelines, and inputs—and should have different expectations for what a qualified output looks like. This oversight can cause a school to make unreasonable promises to the incoming students, which is really a recipe for public disappointment.

If we take a “pathway as the product” approach, we might start to notice that each pathway requires the same amount of iteration as any new product. Also, a pathways approach shows us that while a school’s special sauce can add to the overall quality of the paths, fundamentally a path is an operational process run by humans, and its quality is distinctly tied to how well the humans run it.

Again this is why early success for a bootcamp can be misleading, an educational path works because of the humans on it and replicating the humans isn’t technologically possible yet.

However, a school can increase the number and reliability of its paths in a few ways. By slow-growing them internally, with small teams and cohorts to allow for quick iteration and decision making with a single product owner who is responsible for the success of their individual pathway. However, the quickest way to scale is just to buy another pathway that has PMF already in a single offering. This comes with other challenges, but the reliability of the pathway’s PMF is worth it.

So what is the value of a school anyways?

Fundamentally, a school’s ability to control the day-to-day happenings across paths is very limited, it’s relying on high-quality humans to do what’s best for the students. It’s real core value beyond basic administration and investing resources wisely, is mostly as an attraction mechanism; i..e can the school attract high-quality candidates in large numbers?

The value of the school is the signal. If a school can achieve signal status, then the probability of each individual student's success will also go up. The market will demand more of them, and a community will form that can network in favors.

Beyond signal though, the physics of the program design rule, and as we’ll see in the next post, these mechanisms are far more relevant than any other factor to a school’s success.

Hopefully, this has been a helpful lens upon which to view schools, and you can now see why they can’t have product-market fit and must rely on collecting pathways— aka “courses”— that do. 


Next Post: (Maybe) Responsive Learning using Spaced Repetition

Next in this Series: The Physics of Program Design or “Catapulting Learners to Success”.

Zoom is Powering the Next Wave of Online Learning.

944 Words on Zoom and Learning

Image result for zoom

Zoom is the proverbial steel railway of the online education market, and I’m hard-pressed to find many successful ed-tech companies that aren’t using it in some way to educate their student population. It’s now the default for most human to human instructional purposes, and much like what YouTube was for pre-recorded video, Zoom is for this next generation of ed-tech. 

What are seemingly simple feature improvements to standard video conferencing (stability, recording, remote screen control & whiteboards) are to many instructors a feature-rich tool and the low cost has enabled system-level changes in the ability to scale quality instruction. 

Prior to 2014 at least, video conferencing was simply an unreliable portal to another person’s face, and using online video conferencing felt like teaching with handcuffs. At the time, most online classrooms based on video conferencing tended to feel more like tech support with grandma than anything resembling a rich learning environment. 

But Zoom’s reliability and broad feature set are a much needed boon for instructors who need a platform to effectively interact, and not simply present, to students online. Zoom has therefore enabled new and diverse instructional methodologies that were previously difficult to deliver. 

Improving the Quality of a Process

“I know it sounds crazy to gush over whiteboards, but ask yourself: what would Khan Academy be without whiteboarding?”

back to the future lightning GIF

For many educators, Zoom is the go-to tool for teaching and working with students online. Previously mentioned features like recording sessions allow a student to rewatch key points, breakout rooms enable instructors to group students for projects or 1:1 coaching, and the ability to remotely take control of a student’s system is a massive improvement over the previous generation of just telling someone to “click there, no near the red circle..” that I dare near fainted when I saw it.

Even simple things like built-in whiteboards are huge. I know it sounds crazy to gush over whiteboards, but ask yourself: what would Khan Academy be without whiteboarding? 

Often I’ve recognized that by drastically improving the output or quality of some important processes you can generate system-level improvements. Think of what customer support did for Zappos, or what the George Foreman grill did for my popularity freshman year. 

My favorite example of an improved process overhaul actually occurred during the second Industrial Revolution when we moved from iron to steel railways. Steel lasted longer than iron, and allowed trains to carry far heavier loads, which lead to a positive feedback loop where trains were able to carry more rails and more workers to build more railways to more places.

Zoom has lowered the cost (both in dollars and time) to access a quality instructor and broadened its feature set to include tools that now allow instructors to do the majority of what was previously only possible in a classroom. If more tasks that were once “classroom only” can now be done online, this means more students can access better teachers in a rich learning environment than ever before.

The Creator / Passion Economy

The creator/passion economy is leveraging Zoom along with a suite of tools such as Teachable, Podia, Slack and Zapier to create personal academies. These creators are using Zoom to hold the wide variety of educational experiences mentioned above as well as provide 1:1 coaching on top of their curriculum. 

For instance, Tiago Forte leads a movement of what he calls “Building a Second Brain” and he uses Zoom to maintain contact with his students and hold coaching sessions with groups to ensure that they’re following the methodology correctly. 

This new ability is very important. A student can now instantly tap someone else to get an explanation, help and just emotional support. And teachers like Tiago are taking full advantage of it to hold discussions, seminars, and give 1:1 assistance ON TOP OF a video course on Teachable.

That’s a drastic increase in value to all of his learners.

These new capabilities have also been an obvious boon to bootcamps, as they augment or completely move their physical campuses to online. Most current bootcamps currently have a student size that exceeds those of last generation, and they’re scaling by piggybacking on Zoom’s features. 

And new types of online educational experiences are also popping up on Zoom. In the last month, I’ve attended multiple live virtual lectures with subject matter experts, online seminars with breakout rooms for different discussions, and a whole mini-conference.

Welcome “Consultative Instructors” a.k.a Coaches.

Image result for welcome to the party pal

However, what I’m most excited about is that stable, cheap and feature-rich conferencing provides a unique opening for a different type of instructor to enter the market: the explainers. Folks who are good at “consultative instruction''. 

The difference here in instructional style is subtle, but it would be akin to what I imagine is true of priests (you know, those robed and/or cassocked fellows). Some priests are (probably) better preachers than others; inspiring you from the lectern to new thoughts and regaling you with relatable stories. Others are better in the confessional, helping you with your personal struggles. The same is true of teachers. 

I’ve found that the most productive use of Zoom is in 1:1 consulting, and currently, I’m using Zoom to work with three different coaches to improve my Javascript and writing abilities and to learn Webflow. The experience has been everything I would hope for from an education: it’s personalized, convenient, and cost-effective.

And this is my big hope for the next phase on online learning: access to cheap, personalized instruction while sitting on my couch. Loneliness gone. The cost is reasonable. I can move at my own pace, guided by another human who is actively involved in my educational process. 

Loading more posts…