Coverage raises $42M Series B led by Sequoia and Keith Rabois to disrupt insurance brokerage with flat-fee model

Summary

Coverage, a New York-based insurance brokerage startup, has raised a $42 million Series B co-led by Sequoia and Keith Rabois — the first time Rabois and Sequoia have co-led a deal together since PayPal. Co-founder JD presented the news live, with the round earmarked for product development, sales, and growth rather than reserves. Coverage bears no underwriting risk.

The model

The core bet is a flat-fee structure that replaces the traditional commission-based brokerage model. Legacy brokers — Marsh, Aon, and the other $50–100 billion incumbents — earn more the more a client spends on premiums, which misaligns their incentives with the client's. Coverage charges based on client complexity and servicing needs, then nets out carrier commissions from the client's policies as far as legally possible. The result, JD argues, is genuine alignment: Coverage's job is to get clients the cheapest possible insurance for the best coverage, and carrying its own book would destroy that.

Over 700 companies have moved to the platform, including Gopuff, Eight Sleep, Chomps, and Bombas.

Unit economics

Coverage was profitable in its most recent fiscal year and had not touched its Series A capital when the Series B closed — an unusual position for a growth-stage startup. JD says the challenge now is finding ways to deploy the new capital.

AI impact

AI has tripled per-employee email output over the past six months. JD frames this as breaking the traditional scaling laws of financial services: the business is becoming less reliant on headcount for day-to-day operations, with people redirected toward high-end problem solving. He's direct about the implication — entry-level paper-pushing roles are increasingly hard to justify.

Geography

The company is headquartered in New York, which JD describes as the right base for applied AI in financial services. The strategy is direct competition with the large incumbents — going after their clients rather than selling software to them — and New York is where those clients are.

AI liability as an emerging insurance category

On the question of whether novel AI risks — copyright litigation, model liability, data scraping settlements — are insurable, JD points to Coalition as the template. Coalition entered cyber insurance when no legacy carrier understood the exposure, built it into a $5 billion-plus business, and has since expanded into D&O and other lines. New carriers are now stepping in to underwrite AI-specific risks in the same way, and Coverage positions itself to connect clients with those carriers as the category develops.