What Collapsed in Indian EdTech After 2021

Executive Summary

EdTech demand gravity describes the failure of India’s pandemic-era, paid K–12 replacement model—not the disappearance of digital learning. School closures created emergency usage that several platforms treated as durable, high-value demand. When classrooms reopened, digital tools remained widely used, but increasingly as free or supplementary services rather than recurring paid subscriptions.

Five findings define the correction:

1. Indian EdTech funding fell from $4.6 billion in 2021 to $568 million in 2024; certification and test preparation captured 75% of 2024 funding.
2. The 2023 Bharat Survey for EdTech found 51% current users, 8% discontinued users, and 41% never-users. The original 41% discontinuation claim was incorrect.
3. By 2025, 63% of surveyed children used EdTech, yet only 6% used specialised applications; YouTube, WhatsApp, and Google dominated.
4. Scale did not guarantee viable economics: Byju’s reported a ₹8,245 crore FY22 loss, while Unacademy reported a ₹631 crore FY24 loss.
5. The emerging model is focused and hybrid: PhysicsWallah’s FY26 revenue reached ₹3,900 crore, with online scale supported by 353 offline centres.

The structural error was mistaking access and usage for monetisable, retained demand.

1. What Actually Collapsed

The post-2021 correction did not eliminate digital learning in India. It exposed the weakness of a specific business model: paid K–12 platforms that treated pandemic usage as permanent demand and built large sales, marketing, content, and acquisition organisations around that assumption. During school closures, online learning was a necessary substitute. After reopening, it became a supplement competing with schools, private tuition, coaching centres, and free digital content. Demand gravity is the gap between widespread educational technology usage and the smaller pool of users willing to pay repeatedly for a specialised platform.

2. Emergency Demand and Market Normalisation

Funding shows the scale of the expectation reset. Inc42 recorded $4.6 billion of Indian EdTech funding in 2021, compared with $568 million in 2024. The recovery was concentrated rather than broad: online certification received $216 million and test preparation $212 million, together accounting for 75% of sector funding. K–12 received $53 million. Capital did not leave education completely; it moved toward categories with clearer outcomes, shorter purchase decisions, and more defensible willingness to pay.

The 2023 Bharat Survey for EdTech covered 6,030 low-income households and 9,867 children across six states. It found that 51% of children were current EdTech users, 8% had discontinued, and 41% had never used it. Among children not using EdTech at the time, 61% cited school reopening and 30% cited other learning avenues. This supports a reopening correction, but not a disappearance of digital learning.

3. Usage Persisted, Commercial Capture Weakened

BaSE 2025, covering 12,500 households and 2,500 teachers across ten states, found that 63% of children used EdTech and only 5% had discontinued. However, usage was concentrated in general-purpose tools: 94% used YouTube, 67% WhatsApp, 49% Google, and only 6% used specialised EdTech applications. Seventy-seven percent used technology for practice or doubt resolution, while 47% used it for examination preparation.

The distinction is central. Technology remained embedded in learning, but the dominant mode became low-cost, fragmented, and supplementary. High usage therefore did not automatically produce subscription revenue, long customer lifetimes, or lower acquisition costs for commercial platforms.

4. Scale Did Not Repair the Economics

Company financials show the consequence. Byju’s parent reported FY22 operating revenue of ₹5,014 crore and a consolidated loss of ₹8,245 crore. Unacademy reported FY24 operating revenue of ₹840 crore and a ₹631 crore net loss, even after reducing losses by 62%. These figures do not prove one universal CAC ratio, but they show that scale alone did not absorb the operating structures created during the boom.

Trust increased the cost of retention. In a 2023 LocalCircles online survey, 81% of surveyed platform users reported at least one issue over the previous 24 months; 65% cited refunds, 62% teaching changes or effectiveness, and 35% false promises or transparency. The sample was self-selected, so it should not be treated as national incidence. It still identifies recurring commercial friction.

5. The Surviving Model Is Focused and Hybrid

Offline education was not automatically efficient. Allen reported FY25 operating revenue of ₹3,067 crore, but profit after tax fell to ₹41 crore from ₹136 crore. Its resilience came from established test-preparation demand, local reputation, and classroom economics, not from immunity to cost pressure.

PhysicsWallah illustrates the emerging equilibrium. In FY26 it reported ₹3,900 crore in revenue, 5.34 million paid users, ₹549 crore EBITDA, and a ₹24 crore net loss. Online users reached 4.87 million, while 470,000 offline enrolments generated 45% of revenue through 353 centres. The model combines inexpensive digital distribution with physical support and outcome-oriented categories.

Conclusion

Indian EdTech did not collapse because students rejected technology. The pandemic-era replacement thesis collapsed because access was mistaken for paid, retained demand. Sustainable platforms now require narrower segments, measurable outcomes, transparent pricing, lower-cost distribution, trust, and channel-level economics. Demand gravity remains structural, but execution determines which companies adapt to it.