Education is essential. It’s somewhere between food and roads on the list of essential services (less essential than food, more essential than roads). The American educational system is pretty good, but has needed a redesign for quite a while, and, as with so many other things, covid-19 has shone a spotlight on both the needs and the problems.

Covid-19 also gives us a perfect opportunity to redesign how we provide education, from early childhood through advanced degrees.

Everything is figureoutable.

As an innovator, I need to make clear right up front that it’s insane to think that we can assume and build the perfect answer within a couple months. All the affected parties need to understand and accept that the only way to do this is to work together, try things, stay flexible, and assume positive intent. (Or, I guess I should esay, “assume positive intent with the suggestions for new ways to provide education.” I’m not sure that “pretend like everything’s fine and cram thousands of kids back together” is positive intent. At best, it’s lazy.)

As an experienced innovator, I also know that getting a large group of humans to do those things is going to be far harder than actually creating a great, safe, educational system. But let me lay out my plan and see if there’s any part of it you can pick up and run with.

A note on formatting….

This is a really long and involved essay, but I really want it to start a conversation and maybe prompt action, so I’ve marked places in it with icons from The Noun Project:

This is an action you can do right now
question iconThis is a question or request for input
 idea icon This is a BIG IDEA

My plan

I think this plan, which I’m calling “schoolhoming,” addresses some of the main issues with education in general and education during pandemics in particular. Please let me know what you think is good about it, what’s problematic about it, or any ideas you have for improving it.

What I don’t want to hear is “that will never work because it’s not how we do things now.” I know it isn’t, but we created the system we have now…we can create a different one.

I also don’t want to hear that we can’t do it because it’s hard. We put people on the Moon. This is the future of our nation and possibly our planet; it’s worth a little effort.

idea iconWe need to stop collecting hundreds or thousands of kids and adults in large enclosed industrial buildings. Instead, I suggest that we establish many small “schoolhomes” or school houses throughout each district. We group students into cohorts of about 12, and each cohort has their own location where they get all their schooling. I believe this plan combines the best of what we know about homeschooling, student-teacher ratios, group sizes, psychology, and pedagogy. It also addresses epidemiology, as the cohort functions like a macro-organism, just like the family or care home residents that have been quarantined together.

Belief: Chunking out school into geographically distributed small cohorts is better than the large industrial processing model we use now.

There are several testable hypotheses from this belief. For instance, my school district has offered families the choice of 100% online education or 2 days per week in-person. We could offer a third option of schoolhoming and measure how many families sign up for that. That would give us a go/no-go. If it’s a go (and my prediction is that demand will exceed supply), we select families for the pilot year and provide the education this way, at the elementary, middle school, and high school levels. We then measure the academic achievement, psychological well-being, and rate of infection of the students, teachers, staff, and family members of the whole district and compare the 3 approaches. We also measure the cost of each approach for a cost-benefit analysis.

UPDATE: I go into more detail about cohorts (or pods) here.

I have to hope that school districts are planning to measure all those things this year already, and compare the fully online approach to the in-person approach for a risk/benefit analysis, but I have to admit I haven’t heard anything to suggest they are. If you are a teacher, school administrator, student, or have a kid in school, you might contact your school board and ask them to collect actual data.

So how would this work?

This is obviously a huge change from how we do things now. Public schools are set up to benefit the stakeholders rather then the users – they are as easy and cheap to administer as possible. Everything else is the best you can do within those limits. This is why they are big central buildings: easy to clean, easy to run the physical plant, easy to provide food, easy to transport students to and from.

question iconThe benefits schools have over homeschooling or small private schools is that you have a large and diverse student body so kids get exposed to people and cultures they wouldn’t otherwise, and you can have sports teams and musical groups and interest groups because the small percentage interested in an activity is still large enough to do it. I haven’t figured how to get these benefits with Covid-19, so if you have ideas, please share!

More teachers

If we limit cohort size to 12 students, we would need about one teacher for every 12 students. That is a wonderful ratio for learning! I say “about,” because at the elementary level, the teacher would be part of the cohort and remain with the students from kindergarten through 5th or 6th grade (there are discussions about when to shift kids to specialized learning). Let’s say 6th grade. From 7th to 12th grade, however, each subject should be taught by a subject matter expert, so while the student cohort stays together, teachers rotate between cohorts. We’d need to do the math on how many of which expertise we’d need.

Either way, we need a LOT more teachers.

Turns out there’s a high level of unemployment in the US right now, so that might help…but we should still take a hard look at the incentives for teachers. First up, we need to increase the pay scale for them, and it needs to reflect how much they actually work. We need to fund schools so that teachers don’t have to purchase supplies and furnishings. We need to provide strong health and pension benefits. We should get creative with other benefits – the psychic income is great (I’ve been a high school teacher), but it would be nice to get other bennies.

One thing we might consider is counting teaching as national service, along the lines of military service. Currently education is funded locally, which makes sense, but federal assistance for training and certifying teachers as well as paying for relocation if needed would help. Teachers should get the same corporate love that uniformed military does: preferential seating on airplanes, private lounges, discounts for goods and services, things like that. Our nation has no future without teachers; they are national security.

More buildings

Obviously, right? This is where we will need to get super-creative in the short term. The key is to locate each schoolhome near where the cohort lives if possible, so they don’t have to travel much (to reduce risk, contagion, and time loss). If a family is already homeschooling, they might consider opening their home to more students and that home becomes the schoolhome/cohort. (I’m working with some medical folks on “How to De-Distance Safely”). Otherwise, we need to find spaces that are in the cohort’s neighborhood, in the school district, or at least very close to the school district.

In the long term we might want to build schoolhomes that are specifically designed for the purpose, but for the 2020-2021 school year we should have open minds about what can be a school. Here are some of the first options that come to my mind about finding space for groups of 12+ kids and a couple adults. In all cases we need to figure out custodial services (enhanced ones, for covid). School meals can be provided the way they are now in many districts, with busses running the bus routes to deliver them, but some schools may develop the Cupertino model where parents bring hot lunches to the school. Some neighborhood groups may pull together to provide breakfast and snacks.

Empty stores/strip mall pads and empty mall spaces – these are common, and becoming more common by the day, in suburban and some urban areas.

Pro: they already have electricity, HVAC, WiFi, and maybe even plumbing. They have parking. Pre-schools and commercial “supplemental learning franchises” like Kumon and Mathnasium use these kinds of commercial spaces, so we can use them as models (both for space design and for insurance/legal/zoning).

Con: Landlords need rent and may or may not see their way clear to discounting rent for schools (see the section below on “Convincing Business to Help.”). Not a biolphilic learning environment; little to no nature or outdoors. Not usually an option in dense urban areas.

Airbnbs – Owners of rental vacation properties are being hit hard by the recession and the pandemic. Perhaps some would be happy to host school in their property?

Pro: It’s a win/win for owners who are parents. The location might be in the middle of the residential area where the cohort lives, reducing commute distance. Already have bathrooms, electricity, HVAC, maybe WiFi. A home environment is nice for learning. We could use the same legal/zoning cover as homeschools. May have yard or outdoor area. Apartments may have existing building security.

Con: Only full-unit rentals are a good match. Insurance is a big question. Not a long-term solution in most cases.

Empty office building spaces and co-working spaces – WeWork needs a PR boost, and all coworking spaces are feeling the financial pinch of the pandemic. Office buildings in urban and suburban areas are rarely at 100% capacity. The business owners have a strong motivation to support this, because parents aren’t coming back to work without school options.

Pro: Coworking spaces are ideal for learning/education, especially for junior high and high school. Already finished space. Even unfinished space in office buildings has power and water. Parking is available and most are near public transportation. The buildings have other amenities such as food options. Schools have been housed in office buildings in cities since the early 1900s. Existing building security.

Con: No outdoors/running space in most cases.

Trailers/containers (good enough for our government officers) – More than a few of my friends aned family have lived and worked in converted shipping containers, in the US and in warzones. It’s not ideal, but the USG (especially the DOD) has the acquisition plan in place for these. The standard shipping container is too small for a cohort of 12, but pre-made portable buildings already exist as temporary classrooms and office space. And really, nothing says zombie apocalypse quite like this, right?

Pro: there’s already a whole market segment for portable school buildings, designed with the stuff you need.

Con: we’d need to find places to locate them. The whole point of schoolhoming is to keep cohorts together and not mixed with a larger population, and to distribute the school population, so parking a bunch of trailers on school grounds is exactly the wrong answer. Could we find enough vacant lots, unused parking lots, etc?

Let’s get creative – Schoolhoming is more artisanal than industrial. The important thing is that the physical structure(s) fit the area and the needs of the cohort. At the elementary level this can be a general classroom, but at the high school level we’ll need the space to be a science lab in addition to a standard classroom. 


I know some parents are probably wondering about security against school shooters and other assholes. One of the benefits of distributing a student population like this is you take away the target for school shooters. If they come from outside the school, they are looking for a big dramatic target environment. If they come from inside the cohort…well, that’s much less likely to even get to that point with a cohort model and a low student:teacher ratio. Some of the space options, like office buildings, come with security. For others, we’d want to consider hiring security, and what that should look like given the environment of the schoolhome. Without the large industrialized school, there’s rarely going to be a need for large industrialized security.

Support staff

Hey, more jobs! And possibly closer to home.

One of the things that has been a pain point in the American system for years is that the school day doesn’t match the work day. While it would be nice if employers would subsidize child care, since they expect employees to be at work for 8 or more hours a day (whether they need them or not), let’s assume that’s not going to change by this fall.

What is likely to change is that employers will want employees to start coming back into the office, or to be trackably online from the home office for 8 straight hours…which conflicts with getting a kid to and from school and taking care of that kid when it’s at home. With schoolhoming, additional adult staff could provide childcare before and after classtime at the schoolhome, further reducing US unemployment numbers.

Bussing could shift to vanpooling or carpooling. We don’t want lots of students on busses for contagion reasons, but family members are part of the cohort’s virus organism, so those family members who have a schedule that allows them to transport the kids should do so. Ideally each schoolhome would be within walking distance of the student’s home, but that’s not going to be achievable so some schoolhome will require transport. Partnership with Lyft or another rideshare company could help.

Contracts with local cleaning services or the building’s janitorial service help grow the economy.

Parents or school staff can be rotating IT support.

For middle and high school, where subject teachers move between schoolhomes, plexiglass barriers and other distancing measures will be needed. These teachers should also have their travel reimbursed. Some classes can be taught remotely, with the students attending from the schoolhome and the teacher video-ing in.

Flipped classrooms

One of the problems with online classes as they were done at the end of the 2020 school year is that the instructional design assumed traditional in-person, and that doesn’t translate to online. A flipped classroom model asks the students to read the topic, watch the lecture, and uptake any other foundational instruction on their own, then, when the teacher is available, do the “homework.” (Traditional instruction is that the teacher provides the foundationals then you are on your own for homework). A cohort model would allow either approach as needed; some students will do better struggling to figure out the work on their own for some subjects; at other times it’s easy to watch a video about the basics and then work through it when help is available.

One Size Never Fits All

Just like our industrial education solution is a pretty poor fit for most students, we shouldn’t be overly prescriptive or mandate what each schoolhome is like. Some cohorts will be super-advanced students, others will need special accomodations. Some will be in parks in wealthy areas, others will be in empty storefronts in sketchy areas. Each community and neighborhood needs to come together to make a safe, good, space for students and teachers.

Convincing Businesses to Help

The burden of providing education during a pandemic can’t fall completely on schools. Life is too interdependent for that. If businesses don’t participate, parents will be ground up in the friction between the two systems, and childless employees will be forced to pick up the slack at work. Everyone will be even more stressed and miserable and productivity will decrease.

Many large companies already provide some day care or other aids; rather than having to take on all the support and insurance and everything, they could help financially offset the cost of the schoolhomes. This could take the form of providing space for a schoolhome in their office building (after all, if the parents are already working together the kids are a biological cohort already), contributing furnishing or technology (tax writeoff), donating teacher and staff benefits, etc. If a company that’s already providing day care instead gives the space and the amount of money they are spending on it to their local school district for schoolhoming, it will go much farther.

Smaller businesses are similarly impacted and could team up to help with rent offsets or material contributions. Restaurants could help donate meals, getting advertising and good will as well as the knowledge that their employees won’t quit or call it sick because they have no childcare options.

Essentially, childhood education and childcare is everyone’s business in any community, even for people who don’t have children. If you want a strong economy and a nation that continues to develop, you need to figure out how to provide some element of it. If you don’t want to be the one forced to come into work during a pandemic because the people with kids can’t, you need to help your business figure out how to help.

The data and tracking of all this will be a huge lift, so data science and AI companies and tech companies can help! Designers and architects can help.

Your Turn

What have I not thought of (good or bad)? How might we try this on a small scale in certain places? If you homeschool, can you contribute your best practices or advice? If you are a public school teacher, would you be more or less comfortable with reporting to a small home or office every day, where you and 12-25 students were the only people?

What does a rough-and-ready prototype of this look like in your neighborhood?

We can figure this out!

Chunking out school into geographically distributed small cohorts is better than the large industrial processing model we use now…let’s try it this year!


  1. Thinking about this solely from a pandemic perspective: Is every kid in a schoolhome for their age? So if I have 3 kids of different ages, theyre in 3 different schoolhomes? Now multiply that by every family with 2+ kids. Now imagine just one parent is an ER doc, or a bus driver, or a whatever, who needs child care, so they send their kids to school too. By the time I know an outbreak is happening in the schools (14 days after initial contact), have we just exposed the entire town to COVID?

    1. I’m also struggling to think thru the socio-economic implications. Rich white kids only in school with rich white kids; and so on. Then I imagine the families who can afford it, paying extra for a “better” teacher because it’s still cheaper than private school. Now we created a new marketplace for teachers. Have we just relegated poor, minority students to worse education than they have now? Or maybe not worse… but not better either.

      And this system only exasperates the challenges for families with special needs. Those services (including speech therapy, occupational therapy, counseling, and others) can be pooled because a small staff can (theoretically) serve an entire school. The system is terrible today, I’m not defending it. But how do we distribute it across so many new locations when so many students in special ed programs need access to *multiple* services?

      There are reasons why education is a hard problem to fix in America. It’s not simple or straightforward in any regard!

      1. Right! It’s very complex.

        What about combining special needs students into a cohort? I recognize that might conflict with keeping families together.

      2. I’ve been struggling with the socio-economic issues as well…you make solid points. Initially I think we have to keep kids together that would be hanging out together anyway, but someone just messaged me a suggestion to have cohorts work with each other virtually, so different races and economic groups would interact that way.

        Teacher incentives are a key element.

        Thanks for thinking on this!

  2. I was thinking that cohorts would be families, and then maybe neighbors or friends – whoever would normally be hanging out together. With small class sizes each kid could work at their own level so you could combine ages in one cohort.

  3. I absolutely love the idea of smaller classes AND more broadly, starting a conversation that rethinks the whole education system rather than simply trying to fit round-peg new ideas in the conventional square-peg holes. Some initial thoughts that need further exploration/examination:

    Teachers: Yes, you are right to say we’d need a lot more… but, thinking through a large school district, like Fairfax Co, Virginia — to marry up one teacher to a cohort of 12, you’d have to increase teaching staff by around 25% — add to that the benefits you’ve suggested –that’s not just a lot more — that’s a budgetary nightmare.

    In terms of benefits, while from a theoretical perspective I do not disagree that teachers are deserving much more than they get — however, the increase in pensions also adds to a county’s budgetary burden. While pensions are a great incentive to get teachers in… the other end can prove problematic when the county may be left funding the pensions of hundreds of retired teachers, i.e. those no longer directly contributing but still pulling from an already strapped-for-cash education system, for 20-30 years per teacher.

    Others have questioned how you ensure a more diverse cohort… and that’s a spot on concern. In addition, how do you factor in (also for Fairfax County, Virginia) the 30% who are English-learning students, 30% economically-disadvantaged (an undefined county term – but does that mean they receive free breakfast/lunches — how does that factor in), and then 14.5% students with disabilities. While none of this is insurmountable, it does really complicate the heck out of how to execute such a monumental albeit well-intentioned shift.

    This reads like a nay-sayer and I don’t mean it that way… there are just so many complicating factors to add to the examination… so don’t stop!

    1. Yeah, I figure this is about 12% of a plan…but I’m hoping it prompts folks to think beyond “school like we’ve been doing” or “no school.”

      One option might be to have the students be in cohorts and the teachers video in to multiple cohorts at the same time, both to multiply the reach of the teacher and reduce exposure. You lose a lot that way but it’s better than nothing.

      Students with disabilities could be better served by this. I have no idea how Deaf and hard-of-hearing students are getting online education and I imagine it’s also challenging for blind students. I’ve been using the captioning in Google Meet for online classes (I run workshops for adults) and it’s not awful, but much like screen readers it can cause more confusion than it alleviates.

      It is a budgetary nightmare, which is why I believe it must be done in partnership with the commercial sector. Education doesn’t exist in a vacuum separate from business (just like public health and the economy are not two separate things).

      I’m interested in ways to prototype or pilot new approaches on a small scale, and cheaply, to validate or invalidate them. Fully pivoting to something like this would add too much effort, and not be smart before it’s been tested. I’m also interested in other approaches that address the biggest aspects of the problem…what might work?

    1. Yes it is! I wanted to get it out in front of people quickly to learn whether or not I should spend any more effort on it. I’m hoping that people who like it may see parts of it that they know how to flesh out.

  4. One issue I keep coming back to is how many issues are tied to land use restrictions:
    -socioeconomic diversity in cohorts
    -finding close locations for the new classrooms
    -funding between districts

    1. Agree. While there’s a lot that can be done at a grass-roots level, this is a systems-level change, and as such would require multiple lines of effort. For instance, some amount of diversity could be incorporated by having cohorts interact virtually with each other, partner up kids across cohorts to work on projects together, etc.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.