Embedded Insurance in India Has an API Problem. What It Actually Needs Is a Risk Layer.

Article Summary

This article by Bhakti Dama, Founder of AVYA, argues that India's embedded insurance has scaled distribution while failing to close the protection gap. Although the country issued 41.84 crore policies in FY 2024–25 and the embedded market is on track to cross USD 2 billion by 2026, insurance penetration fell to 3.7% of GDP in FY25. Dama contends the industry has confused distribution with diagnosis, and proposes a “risk layer” that assesses each customer's real exposure before any product is offered. Readers will understand why risk identification, not better APIs, is insurance's next competitive moat.

India's embedded insurance market is growing three times faster than standalone insurance. Yet the national protection gap keeps widening. The reason: we have confused distribution with diagnosis.

By Bhakti Dama, Founder — AVYA | June 2026

Here is a question worth sitting with: India issued 41.84 crore insurance policies in FY 2024–25. The embedded insurance market is on track to cross USD 2 billion by 2026. Embedded insurance is growing three times faster than standalone. And yet — India's insurance penetration declined to 3.7% of GDP in FY25, the third consecutive year of decline from its pandemic-era peak.

How is it possible to sell more insurance than ever before, while protection actually diminishes?

The answer lies in a distinction the industry has not made clearly enough: there is a profound difference between distributing insurance and matching someone to the right protection. India's embedded insurance revolution has been almost entirely about the former. It has delivered extraordinary scale and near-zero coverage.

The API Gold Rush — and Its Ceiling

Embedded insurance in India today is largely a product-push architecture. A customer books a flight — travel insurance appears. A smartphone is purchased — device cover is offered. A personal loan is disbursed — credit life is bundled. Each of these is a legitimate product. None of them required the platform to ask: what does this person actually need?

This is the API model of embedded insurance. An insurer or aggregator provides an integration layer. A platform drops insurance into a customer journey at a predefined trigger point. Distribution happens. A policy count increases. The protection gap does not close.

The numbers reveal this tension clearly. India's insurance density — premium spend per capita — stands at just USD 97, against a global average of USD 943. Life insurance penetration has slipped to 2.7% of GDP. Non-life holds at 1%, against a global non-life average of 4.3%. We have scaled distribution. We have not scaled protection.

India has confused the act of selling insurance with the act of protecting people. These are not the same thing.

The embedded insurance market is not failing. It is succeeding at exactly what it was designed to do: reduce friction in a transaction. The problem is that friction reduction was never the same as need fulfilment. When a customer's biggest risk is the cost of a hospitalisation, selling them device insurance at the point of a phone purchase is not protecting them — it is occupying their insurance attention span.

The Awareness Gap Is Not a Communication Problem

Industry discussions about embedded insurance's underperformance frequently arrive at the same diagnosis: consumers are not aware of the coverage they hold. They do not file claims. They do not renew. The proposed fix is always better communication — clearer disclosures, better onboarding, more reminders.

This framing misses something fundamental. A customer who was not aware of their coverage was, in many cases, not a customer with a real need for that product in the first place. The awareness gap is not primarily a communication problem. It is a relevance problem.

When insurance is embedded as an add-on to a transaction rather than a response to an identified risk, it occupies a slot in the customer's financial life without earning it. The customer did not seek it. No one helped them understand whether it addressed anything they were actually worried about. The policy exists. The protection is largely theoretical.

This is structurally different from a customer who was helped to understand their actual risk exposure, discovered a genuine gap, and then chose a product that addressed it. That customer files claims. That customer renews. That customer refers.

Diagnosis Before Prescription: What a Risk Layer Actually Means

In medicine, a prescription without a diagnosis is not medicine — it is guesswork. The pharmaceutical industry does not attempt to distribute drugs through e-commerce platforms without prescription. Yet the insurance industry has built its entire embedded architecture on exactly this model: product distribution without prior risk assessment.

A risk layer changes this sequence. Before a product is offered, the customer's actual exposure is surfaced. What is their income structure — salaried, self-employed, gig? Do they carry dependants? What life stage are they in? What are their existing coverages and where are the gaps? What financial obligations would be most disrupted by a health emergency, an income interruption, or a liability event?

This is not a theoretical exercise. It is the precondition for meaningful coverage. And it is precisely where the current embedded insurance architecture has no infrastructure.

The most underserved insurance customer in India is not someone who has never encountered insurance. It is someone who has encountered insurance many times — bundled into a loan, embedded into a booking, ticked as an add-on — and has never once been asked: what actually puts your financial life at risk?

The Regulatory Signal Is Already Here

India's regulatory evolution is pointing in exactly this direction, even if the industry has not fully read the signal.

The Sabka Bima Sabki Raksha (Amendment of Insurance Laws) Act, 2025, passed by Parliament in December 2025, formally recognises Managing General Agents (MGAs) as a class of insurance intermediary. This is not a technical footnote. MGAs, by design, are specialised underwriting and distribution entities with delegated authority from insurers — the ability to bind risk, set pricing within parameters, and design products for specific segments. An MGA is, structurally, a risk identification and product-matching entity, not a distribution pipe.

The formalisation of the MGA framework is IRDAI's implicit acknowledgment that the next phase of insurance in India requires more sophisticated intermediaries — ones that can operate at the intersection of risk knowledge and product design, not just at the point of transaction.

Simultaneously, the Bima Trinity — Bima Sugam, Bima Vahak, and Bima Vistaar — is creating distribution infrastructure at a scale that only makes sense if what is being distributed is contextually relevant. Bima Vahak, in particular, positions women as the delivery channel for insurance in sub-urban and rural India, specifically because women who understand the community's needs are better positioned to identify risk gaps than any algorithmic trigger on a transaction platform.

IRDAI's Vision 2047, with its goal of doubling insurance penetration to 8% of GDP, cannot be achieved through product-push embedded insurance alone. The mathematics do not work. Every percentage point of penetration growth requires reaching a customer who is not already covered, with a product that addresses something real in their life. That requires risk identification as a precondition to distribution.

What the Next Architecture Looks Like

The embedded insurance model of the next decade will be distinguished not by how seamlessly a product can be pushed into a customer journey, but by whether a platform can answer, before the push, whether that product is the right one for that customer.

This requires three capabilities that today's embedded insurance infrastructure largely lacks.

The first is a risk diagnostic layer — a mechanism, embedded in the platform journey, that surfaces the customer's actual financial exposure before any product is offered. This does not need to be a 40-question survey. A five-minute structured needs assessment, contextualised to the platform's customer type, can reveal the most material gaps.

The second is a needs-to-product matching logic — the ability to map what the diagnostic reveals to the specific products in the panel that address those gaps most directly. This is where the risk layer creates commercial advantage: a platform that recommends the right cover earns trust; a platform that pushes an irrelevant one burns it.

The third is consent-based, opt-in distribution — presenting coverage as a response to a surfaced need, not as an add-on to a transaction. When a customer understands why a product is being offered and what risk it addresses, uptake, claim rates, and renewal all improve. The product performs. The unit economics work.

The next moat in Indian insurance is not who has the best API. It is who can answer the question: what does this specific customer actually need to be protected from?

A Call to the Industry

India's insurtechs have done extraordinary work building the distribution infrastructure for embedded insurance. Lowering friction, improving reach, digitising the purchase journey — these are genuine achievements. But distribution infrastructure, on its own, cannot close a protection gap.

The industry's next investment needs to go into the diagnostic layer that sits before distribution. Into risk assessment tools that work at the scale and speed of embedded platforms. Into product-matching logic that is specific to customer segments, life stages, and income profiles. Into intermediary models — MGAs, specialised brokers, context-aware distribution partners — that bring underwriting intelligence into the customer journey, not just processing efficiency.

The question every embedded insurance platform should be asking itself is not: how many policies did we sell? It is: how many of those customers are meaningfully more protected today than they were before we sold them a policy?

India's insurance penetration has declined for three years running. The embedded insurance market is growing. These two facts, held together, tell us everything we need to know about what the next decade requires.

Distribute less. Diagnose more. Match precisely.

About the Author

Bhakti Dama

Bhakti Dama is the Founder of AVYA, India's AI-powered risk management and insurance platform. With 25 years of experience in underwriting and product innovation across Life, Health, and General Insurance, she is the inventor of the patented Try & Buy insurance model — India's first contextual, consent-based insurance distribution framework. She writes on the intersection of risk, product design, and financial inclusion.

Key Highlights

• India's embedded insurance market is growing three times faster than standalone insurance and is on track to cross USD 2 billion by 2026, yet India's insurance penetration declined to 3.7% of GDP in FY25 — a third consecutive year of decline.

• India's embedded insurance shortfall is a relevance problem, not a communication problem: products bundled into transactions occupy a customer's attention without addressing their real risk, which depresses claims and renewals.

• A risk layer — a diagnostic step that surfaces a customer's financial exposure before any product is offered — is the next competitive moat in Indian insurance, reinforced by the 2025 recognition of MGAs and IRDAI's Vision 2047 goal of doubling penetration to 8% of GDP.

  
  

Frequently Asked Questions

What is a “risk layer” in embedded insurance?

A risk layer is a diagnostic mechanism embedded in the platform journey that surfaces a customer's actual financial exposure — income structure, dependants, life stage, and existing coverage gaps — before any insurance product is offered. It replaces product-push distribution with needs-based matching. The author frames it as the difference between diagnosis and prescription: offering coverage as a response to an identified risk rather than as an add-on to a transaction.

Why is India's insurance penetration falling even as embedded insurance grows?

Because India has confused distributing insurance with protecting people. Embedded insurance today is largely product-push — travel cover at flight booking, device cover at phone purchase — which raises policy counts without addressing customers' real risks. India's insurance density is just USD 97 per capita against a global average of USD 943, and penetration fell to 3.7% of GDP in FY25. Scaling distribution has not scaled protection.

How does India's regulation support a shift toward risk-based insurance?

The Sabka Bima Sabki Raksha Act, 2025 formally recognises Managing General Agents (MGAs) as specialised underwriting and product-matching intermediaries rather than mere distribution pipes. The Bima Trinity — Bima Sugam, Bima Vahak, and Bima Vistaar — builds distribution that only works if what is distributed is contextually relevant. IRDAI's Vision 2047 target of 8% penetration cannot be met through product-push alone, which requires risk identification before distribution.

Disclaimer: The opinions expressed within this article are the personal opinions of the author. The facts and opinions appearing in the article do not reflect the views of IIA, and IIA does not assume any responsibility or liability for the same.