No one buys a product they cannot inspect, whose quality reveals itself over a decade, whose switching costs are borne by their own child, and whose sellers all claim excellence. Except that every parent does, and calls it choosing a school. The enrollment decision is one of the largest purchases a family makes, in money or in forgone alternatives, and it is made almost entirely on proxies. Understanding which proxies, and why, explains more about the education market than any league table.
The official story is that parents compare academic quality. The observable behavior says otherwise. Academic quality is nearly uninspectable from outside: results data is confounded by intake, classroom practice is invisible on a tour, and the outcome that matters, who this child becomes, has a ten-year reporting lag. So parents do what all customers do under radical uncertainty: they substitute inspectable signals for uninspectable quality. The peer group, legible in the parking lot in about four minutes. Safety and order, read from corridors and bathrooms. Badges: accreditations, rankings, the crest on the jumper. And the most powerful instrument of all, the child's own face at pickup.
The badge problem
Badges deserve suspicion in this market for the same reason they deserve it in every market: they are sold. Ricardo Semler has told the story of serving on the jury of a best-companies award where the winner was a bank whose tellers were not allowed to sit; the runner-up served kombucha. Culture certifications, he noted, are paid for, the company pays the fee, runs its own survey, and hangs the plaque. Education's badge economy runs on adjacent physics. Rankings reward selectivity, which measures the intake, not the teaching; accreditation measures paperwork and process; awards go to schools with the staff capacity to apply for awards. None of this makes badges worthless, but it makes them what they are: marketing that has passed an audit, not evidence about a particular child's coming decade.
What cuts through, in practice, is the visceral channel. Parents defect from schools when the child's affect changes: the Sunday-evening dread, the stomachaches, the flatness. And they stay loyal, sometimes against test-score evidence, when the child is visibly alive there. The market data from education's newest segment says this plainly. When the National Microschooling Center surveyed 800 American microschools, the outcomes founders ranked as mattering most to their families were growth in nonacademic learning, children's happiness, and skills perceived as needed for the future, alongside academic growth (Soifer and Soifer, 2025). The same survey shows the segment pricing for its actual customer base: 74 percent charge $10,000 or less and 65 percent offer sliding-scale tuition. This is not a luxury market selling prestige. It is a mass market selling reassurance that a specific child is known and okay.
Why bad schools keep their customers
The customer lens also explains the market's most depressing feature: persistently weak schools with stable enrollment. Switching schools is not like switching brands. The cost lands on the child, friendships severed, status reset, and the information problem never improves: the next school is exactly as uninspectable as this one. Loss aversion under uncertainty produces exactly what the system shows, massive incumbency advantage. Most families do not choose a school so much as fail to accumulate sufficient reason to leave one. Exit requires a shock, a bullying incident, a collapse in the child, a move, which is why enrollment change concentrates around crises rather than quality comparisons.
This is also why the fastest-growing alternatives recruit the way they do: not by publishing better scores, but by offering what the incumbent structurally cannot, smallness, a named adult who knows the child, visible daily evidence of the child's state. They compete on the parent's actual decision variables instead of the official ones.
The honest conclusion
It is fashionable to lament that parents choose schools for the wrong reasons. The customer lens suggests the reverse: parents are reasoning correctly about a market that gives them almost nothing real to reason with. A rational buyer facing unverifiable quality claims buys on trust signals and the child's face. If educators want the enrollment decision made on better information, the remedy is not scolding parents for reading proxies. It is producing the one dataset parents cannot get elsewhere and want most: continuous, concrete evidence of what their particular child did, made, struggled with, and became this month. Schools that show the work will beat schools that show the plaque. The customers have been asking for this all along; they simply had no supplier.
Sources
- National Microschooling Center / Soifer, D. & Soifer, A. (2025). American Microschools: A Sector Analysis. 800 microschools: priority outcomes ranked as nonacademic growth, children's happiness, future skills, academic growth; 74% tuition at or below $10,000; 65% sliding scale. https://microschoolingcenter.org/news-blog/2025-american-microschools-sector-analysis
- National Microschooling Center (2024). How many students do microschools serve? Median microschool enrollment of 9 at launch, 12 after year one, 16 at survey; first-year enrollment churn concentrated in autumn months. https://microschoolingcenter.org/news-blog/how-many-students-do-microschools-serve