Stride's $2.5 billion question: who pays for an online school?

If a school moves online, who actually saves money? The parent, the operator, or the government?

I think that is a more useful starting question than whether online schooling can scale. Stride already gives us a sizeable business to examine. The harder question is what that business needs around the screen, and whether the same arrangement would make sense in India.

An online school still needs somebody to pay for the work of schooling. The delivery channel changes. The responsibilities do not disappear.

Read the business before the technology

Stride reported $2.518 billion in revenue for the financial year ended 30 June 2026, up 4.7%. Average enrolment in its full-service school programmes was 243,900, up 4.2%. Its reported annual revenue per enrolment was $9,914. The results were released on 4 August. ^[1]

Those are substantial numbers. But $9,914 is company revenue per enrolment, not a fee paid by every family. Nor should we divide consolidated revenue by those pupils and call the answer school fees: Stride also has an Adult Learning business, which its school enrolment measure excludes. ^[1]

The financing matters. Stride's annual filing says most revenue depends on comprehensive services for publicly funded schools. Public budgets and per-pupil funding formulas are central to the model. ^[2]

So, when an Indian founder sees a tuition-free American online school, the interesting question is who pays its provider. Free at the point of use does not mean the operation runs without a bill.

The screen does not erase the teaching budget

Instructional costs and services were $1.567 billion in FY2026, or 62.2% of revenue. That share was 60.8% a year earlier. Stride identifies hiring, salaries, student computers and technology amortisation among the reasons those costs increased. ^[2]

Put simply, roughly 62 cents of each revenue dollar went into that cost category. I would be careful about treating a virtual school as a content library with an attractive subscription margin. It has ongoing delivery obligations.

GAAP operating income divided by revenue gives a 17.9% operating margin, against 15.0% in FY2025. But the earlier year contained a $59.5 million impairment, while FY2026 contained none. The improvement cannot all be described as a more efficient school operation. ^[2]

The point I take from this is a working hypothesis: the repeatable advantage may sit in coordinating teaching and support reliably across locations. I would want to test that hypothesis against student outcomes, rather than accept it from financial scale alone.

There is another operating budget: the family's time

Indiana Digital Learning School, a K12-powered public school, describes a typical Learning Coach commitment of four to six hours a day for grades K–6. This is the school's guidance, not an independently measured average for all Stride families. ^[3]

Across an illustrative five-day week, that becomes 20 to 30 hours. We should hear the implication: an adult may need to organise a substantial part of the child's day.

I would put that time beside the provider's bill when evaluating an Indian model. If a parent can comfortably provide it, the arrangement might work well. If both parents are unavailable, somebody else must supply the support. A local learning centre could be one answer, but then supervision, premises and staffing enter the budget. That is a proposal to investigate, not a demonstrated saving.

The strongest objection is also useful

Does this mean online schooling cannot be cheaper or better? No. Shared curriculum and technology may reduce some duplication. Families may value flexibility enough to make the adult commitment worthwhile. Some children may benefit from a different environment.

But those possibilities need their own evidence. A company's revenue growth does not tell us how much a particular pupil learned, whether the improvement lasted, or whether a family could sustain the arrangement. I would want comparable outcomes and a clear account of the pupils each model serves.

For an Indian pilot, I would ask three things before making a large claim. Who funds each child's place? Who provides daily support when the child gets stuck? And what result are we buying beyond attendance or completed assignments?

Then I would compare the full cost with a credible alternative. Include the public or private payment, delivery costs and household commitments. Explain which costs are measured and which are assumptions. Keep the outcome definition consistent.

Stride is worth studying because it makes the operating questions visible. My proposed lesson for India is to borrow that discipline: identify the payer, specify the support, and measure the learning. The screen is the route. A working school is the product.