Somewhere between 750,000 and 2.1 million American children now attend a microschool as their main form of schooling, according to the National Microschooling Center's estimates. A range that wide is itself the most honest data point in the sector: nobody actually knows how big this is. There is no federal registry, no common definition, and in many states no reporting requirement at all. What exists is a movement growing faster than anyone's ability to count it.
Start with what is reasonably solid. The Center's 2025 sector analysis surveyed 800 operating and pre-launch microschools across all 50 states and the District of Columbia (Soifer and Soifer, 2025). The schools it describes are small by design and cheap by private-school standards: 74 percent charge annual tuition and fees at or below $10,000, and 65 percent offer sliding-scale tuition or discounts. Earlier sector data gives a sense of scale per school: a median of 9 students at launch, 12 by the end of the first year, 16 at the time of survey. These are not boutique academies for the wealthy. They are, structurally, one-room schoolhouses reborn with a stripe of software.
What the numbers can and cannot say
The outcome claims deserve more caution. Among microschools that track student progress over time, 81 percent reported between one and two years of academic gains in a single school year. That figure comes from the schools themselves, about their own students, with no external verification, no common instrument, and an obvious incentive to report success. It should be read as a claim the sector makes about itself, not as evidence. The fair summary is that microschool academic outcomes are essentially unstudied at scale, and anyone quoting the 81 percent as proof is selling something.
What the survey does establish, because it is descriptive rather than evaluative, is what these schools value. Asked about the outcomes that matter most, founders ranked growth in nonacademic learning, children's happiness, and skills perceived as needed for the future alongside academic growth. And only 29 percent use traditional letter grades; the rest run on observation-based reporting, portfolios, and mastery tracking. Whatever else microschools are, they are a mass experiment in assessment without grading, conducted by hundreds of independent operators who mostly have not read each other's work.
The old logic underneath the new form
The design instinct is older than the funding mechanisms driving the current wave. As Ricardo Semler has told it, two decades of schools where children choose what to learn, with roughly 3,000 children across six countries, ran on the same skeleton many microschools are now rediscovering: no rows, no bells, no grades for memorization, just a project, a tutor, and a deadline, which, as he puts it, is the shape of work too. The microschool surge looks less like an invention than a decentralized rediscovery, this time propelled by pandemic-era exits from district schools and, in many states, by education savings accounts that let public money follow the child into unaccredited settings.
That funding link is also the sector's open question. Policy tailwinds can inflate a movement faster than its quality control matures, and a schooling form with a median enrollment of sixteen has no institutional immune system: one weak operator is the whole school. The sector's own data shows churn in the first year, with enrollment dips in the autumn months as families test and sometimes leave.
Reading the surge correctly
Two readings of the microschool moment are both defensible. One: this is a genuine structural exit, millions of families concluding that the industrial classroom is not worth reforming and building small alternatives with their own hands. Two: this is a loosely counted, lightly regulated boom that will consolidate, professionalize, or partially collapse as the money and the novelty settle.
The evidence supports holding both. The demand is real and the form is coherent; the outcomes are unverified and the counting is soft. What can be said with confidence is narrower and more interesting: a measurable share of American childhood has moved into rooms of a dozen students where nobody issues letter grades. Whether or not the microschool becomes the future of schooling, it has already become the largest live experiment in what school looks like when parents design it from scratch.
Sources
- National Microschooling Center / Soifer, D. & Soifer, A. (2025). American Microschools: A Sector Analysis. 800 microschools surveyed across 50 states + DC; 74% at or below $10,000 tuition; 65% sliding scale; 29% use letter grades; 81% self-reported 1-2 years of gains. https://microschoolingcenter.org/news-blog/2025-american-microschools-sector-analysis
- National Microschooling Center (2024). How many students do microschools serve? Median enrollment 9 at launch, 12 after year one, 16 at survey; national estimate 750,000-2.1 million students. https://microschoolingcenter.org/news-blog/how-many-students-do-microschools-serve