Insurance · Deep dive

Ominimo Insurance

A profitable Hungarian-Serbian MGA that sells AI-priced car insurance on other insurers' balance sheets — 7% of Hungary's motor market in year one, a €1.4B valuation two years after launch, and a plan to take its own licence and cross the Atlantic.

emerging

The question that decides it: Ominimo's loss-ratio edge was proven in Hungary, an underpriced CEE motor market where incumbents rate on a handful of variables and a bootstrapped MGA could undercut them profitably. Does that pricing advantage survive the move into Spain, Italy, France and the US — markets where incumbents already rate on dense data, acquisition costs are multiples higher, and Ominimo plans to swap its capital-light MGA structure for its own licence and balance-sheet risk just as the 92-100% combined ratio has to hold on someone else's regulatory capital becoming its own?

My take ★★ Very interesting ★ Interesting − Maybe ✗ Not interesting

Snapshot

Ominimo sells car insurance priced by an unusually deep data model, and does so profitably — a sentence almost no European insurtech of the last decade could write. Launched in Hungary in 2024, it took roughly 7% of the Hungarian motor market and 300,000+ policies in its first twelve months (TechCrunch, April 2025), expanded to Poland, the Netherlands and Sweden on Zurich’s paper, and grew annualised gross written premium from €26.3M in 2024 to €157.8M in 2025 to roughly €306.8M by July 2026 (company figures). In July 2026 the EBRD’s venture arm led a €20.1M first close of its Series B at €1.4B ($1.6B) — Serbia’s first home-grown unicorn, two years after launch, on barely €30M of outside capital. It is now spending that credibility on its riskiest moves yet: its own insurance licence, Western Europe, and a 2027 US entry.

Founding story

Dusan Komar spent reportedly around twelve years at McKinsey advising European insurers on AI-based pricing and digital business building (fintech.global, April 2025). The recurring failure mode he describes: consultants build a modern pricing product, hand over the code, and watch it die inside rigid legacy systems, slow corporate decision loops and organisations no ambitious engineer wants to join. “No brilliant software engineer or data scientist dreams of working for an insurance company,” as he put it to TechCrunch (April 2025).

With Dennis Weinbender (now chief pricing and data officer) and Laslo Horvath (CTO), Komar inverted the consulting model: build the product as your own company, own the outcome. They launched bootstrapped in Hungary in 2024 — no seed round — and made talent the founding strategy: competitive pay plus intellectual licence for Central and Eastern Europe’s mathematical elite. It worked — the team includes eight Mathematics Olympiad medallists and one Physics Olympiad medallist, and about two-thirds of the ~130 staff work in data science or software (company, July 2026). A fourth co-founder, Kristina Kozina, later joined (Tech Funding News, July 2026). The company is Budapest-registered with its founding and engineering nucleus substantially in Belgrade, which is why both countries claim the unicorn.

How it works

Ominimo is not an insurer. It is a managing general agent (MGA): it designs the product, prices the risk, acquires the customer, issues the policy and handles claims, while a licensed carrier holds the risk and regulatory capital — Signal Iduna in Hungary (policies sell as “Ominimo, powered by SIGNAL IDUNA”), and DA Direkt, Zurich’s German direct unit, in Poland, the Netherlands and Sweden (Coverager; Insurance Business, April 2025).

The claimed edge is the rating engine. A traditional motor insurer prices on five or six main parameters — age, vehicle type, location, driving history; a modern one might use fifteen. Ominimo uses hundreds. Komar’s example: a licence plate unlocks a vehicle database with about 100 variables — length, height, width, weight — and the data shows a strong correlation between car length and parking-accident frequency (TechCrunch, April 2025). Layer in population density and similar non-obvious features and the model separates risks incumbents lump together — quoting good risks cheaper, leaving bad risks to competitors who can’t tell the difference. The company says the portfolio runs at a combined ratio between 92% and 100% even after reserving for ultimate claims liabilities, including RBNS and IBNR provisions (The Recursive, July 2026) — a conservative framing if accurate, with a two-year-old book as the operative caveat.

Distribution is deliberately boring: Hungarian mandatory third-party liability (KGFB) is a comparison-shopped product with an annual switching season — a structurally ideal wedge for whoever is cheapest, which a better model lets you be without losing money.

Product and business overview

The product set is narrow by design: consumer motor insurance — compulsory third-party liability first, in markets where it is mandatory and comparison-shopped — sold direct and through aggregators, on a fully digital quote-bind-claims journey built greenfield by Horvath’s team, with third-party services plugged in via API where building made no sense. Live markets as of July 2026: Hungary (2024, Signal Iduna paper), Poland (Q2 2025, via DA Direkt), the Netherlands and Sweden. Announced next: Belgium and Romania, then Spain, Italy and France, then a US launch targeted for 2027, plus new lines — property has been flagged since the Zurich deal — and Ominimo’s own insurance licence (EU-Startups, July 2026). The licence is the pivotal decision on the roadmap: it converts a capital-light broker into a risk carrier, with everything that implies for capital, regulation and what the equity is worth per euro of premium.

Business model and pricing

As an MGA, Ominimo earns commissions and fees on premium written on partner paper, plus (typically; terms undisclosed) profit-sharing linked to underwriting performance — the structure that lets ~€307M of annualised GWP flow through a 130-person company that has raised only ~€30M. It has been profitable since its first year (company; TechCrunch, April 2025); management says the Series B funds acceleration, not operations (EU-Startups, July 2026).

Consumer pricing is quote-based and aggressively positioned; the model’s whole point is selective cheapness. There is no published rate card, but the Hungarian complaint record shows the mechanics’ sharp edge: one customer reported a renewal quote jumping from 48,786 to 287,386 forints (roughly €125 to €740) with no claims — what re-rating a risk the model no longer wants looks like from outside (gyakorikerdesek.hu, 2025-2026). Valuation math worth stating plainly: €1.4B is roughly 4.6x annualised GWP — premium volume, not revenue; Ominimo’s actual revenue is the undisclosed commission slice. And the EBRD round is small (€20.1M), so the €1.4B print rests on a thin sliver of the cap table.

Traction over time

The GWP curve is the story: roughly 6x in 2025, nearly 2x again by mid-2026, with headcount near 130 — premium per employee north of €2M, absurdly lean for insurance distribution. The 7x valuation step-up in fifteen months outpaced GWP growth (~2x over the window), so multiple expansion — the scarcity value of a profitable insurtech — did most of the work.

Market analysis

European motor insurance is enormous and nearly growthless: roughly $194B in gross written premium in 2025, growing at about a 1.9% CAGR through 2030 (Statista Market Forecast, 2025); broader definitions run to $271B (Expert Market Research, 2025). This is a share-shift game, not a rising tide. The structural forces favour Ominimo’s wedge: mandatory motor liability makes demand universal; aggregator adoption makes price the dominant buying criterion; claims inflation since 2022 has forced incumbent rate increases that widen the arbitrage for anyone pricing more precisely; and CEE combines under-modelled markets with world-class, under-priced technical talent. The US move targets a vastly larger private auto market — but one where GEICO and Progressive spend billions on advertising and telematics-driven pricing, the opposite of under-modelled.

Competitive intel

The set splits into mirrors, monsters and gatekeepers. Prima Assicurazioni is the mirror — the Italian data-driven motor player that sold 51% to AXA in August 2025, validating the playbook and telegraphing the endgame. Marshmallow ($370M revenue 2024, profitable, $2B+ valuation, April 2025) proves a data-edge motor insurtech can win a mature Western market — though by targeting an under-served segment (UK newcomers), not the mainstream. Wefox is the monster-under-the-bed: the $4.5B distribution-first unicorn that nearly died in 2024 and sold its motor book — the contrast that makes Ominimo fundable and the memory that punishes it if loss ratios slip. Lemonade brings brand and a $4B market cap but a decade of underwriting losses; Komar explicitly invites the loss-ratio comparison. The real fight in the next markets is with incumbent direct brands — Linea Directa in Spain, Direct Assurance in France, Genertel in Italy — already cheap, digital and data-rich. And Signal Iduna and DA Direkt are partners who currently control Ominimo’s capacity and could constrain or copy it — precisely why the own-licence plan exists.

History and evolution

• 2023-2024 — Komar, Weinbender and Horvath incorporate and build; bootstrapped Hungary launch in 2024, selling mandatory motor liability on Signal Iduna’s paper.
• Early 2025 — ~300,000 policies and ~7% Hungarian market share inside twelve months; profitable; annualised GWP €26.3M exiting 2024.
• April 2025 — Zurich takes a minority stake — reportedly €10M for ~5% at a €200M ($220M) valuation (TechCrunch) — plus a DA Direkt distribution partnership covering 10+ planned markets.
• Q2 2025 — Poland launch via DA Direkt; Netherlands and Sweden follow in 2025.
• 2025 — Annualised GWP reaches €157.8M; Dutch Trustpilot builds to 4.5/5 across ~231 reviews, with a visible tail of service complaints.
• July 27-28, 2026 — €20.1M ($23M) Series B first close led by EBRD Venture Capital at €1.4B ($1.6B); annualised GWP ~€306.8M; ~130 staff. Plans announced: own licence, Belgium/Romania then Spain/Italy/France, US in 2027, expansion beyond motor.

No major public stumbles yet — which, for a two-year-old insurance book, mostly means it hasn’t aged through enough claims seasons to reveal them.

What people say

The case for. Dutch customers rate Ominimo 4.5/5 across ~231 Trustpilot reviews (accessed July 2026); recurring themes are sharply lower prices than incumbents and fast, friendly service when it works, and the company answers 80% of negative reviews within about two weeks. Hungarian forum threads include multi-year customers reporting zero problems, noting that for a standardized product like KGFB, cheapest-wins is rational (gyakorikerdesek.hu, 2025-2026). Investors supply the loudest praise: EBRD called out the speed to unicorn status versus its prior portfolio; Zurich’s EMEA CEO Alison Martin framed the stake as core to Zurich’s 2025-2027 profitable-retail-growth ambitions.

The complaints. The negative Trustpilot tail is specific and operational: customer service unresponsive for weeks, policy terms changed without the customer’s awareness — including payment frequency switched from monthly to yearly — and documentation that never arrived after a month of calls and emails (Trustpilot NL, accessed July 2026). Hungarian users report payment reminders after on-time payment, automated-only support responses, and the renewal-quote jump noted above (gyakorikerdesek.hu, 2025-2026). Employee-review data is essentially absent — a single 3.0 Glassdoor rating (accessed July 2026) — so the celebrated culture is untestable from outside. Structurally: the combined-ratio and market-share figures are company-reported, not audited; the valuation prices ~4.6x GWP for what is currently a commission business; and the book is too young for its loss picks to have matured — Wefox also looked like a machine at year two.

Outlook: the open question

Ominimo works if the Hungarian result was the model, not the market — if, in Spain, Italy, France and eventually the US, its rating engine still finds mispriced risks that data-rich incumbent direct brands have missed, holds the combined ratio in the self-reported 92-100% band as the book ages through full claims cycles, and converts today’s commission stream into underwriting profit under its own licence without the capital strain that killed most insurtechs that tried it. It stalls if Hungary was an easy first mark — an under-modelled CEE market with crude incumbent rating — and the edge compresses where Linea Directa and Direct Assurance already price on dense data; if acquisition costs or aggregator commoditisation eat the margin; or if the licence transition lands just as loss ratios normalise, leaving a thinly capitalised carrier valued like a software company. The tells: whether the Poland, Netherlands and Sweden cohorts show Hungary-grade loss ratios (not yet broken out); whether an audited combined ratio appears once the licence application forces disclosure; whether Zurich deepens or dilutes — a Prima-style takeout would cap the independent-company thesis; and whether the 2027 US plan survives GEICO-level pricing sophistication or is quietly deferred. The service-complaint pattern matters: in a switching-season product, the model wins the customer, but operations keep them.

How a challenger would attack it

Hit the seams between model and operations. Ominimo’s rating engine is real, but its service layer is visibly thin: weeks-long support silences, policy terms changed without customer awareness, payment frequency switched from monthly to yearly, documentation that never arrives — all on the record in Dutch Trustpilot and Hungarian forums. In a comparison-shopped, annual-switching product, a rival with Ominimo-grade pricing plus actually functioning operations wins the renewal season Ominimo’s model wins the first sale of. The attack is straightforward: replicate the approach — hundreds of rating variables from public vehicle databases is a method, not a moat, and Signal Iduna and DA Direkt see the playbook from inside — then target Ominimo’s own churned customers, starting with the ones re-rated out (the €125-to-€740 renewal jump is a lead-generation list). The deeper vulnerability is timing: Ominimo is swapping capital-light MGA economics for its own licence and balance sheet exactly as it enters Spain, Italy and France, where Linea Directa and Direct Assurance already price on dense data. A challenger that stays MGA — pure commission, no regulatory capital, partner paper — keeps the lean 130-person cost structure Ominimo is about to abandon, and can undercut in the very markets where Ominimo’s Hungarian edge is least proven.

Same playbook, new buyer

Run the under-modelled-market wedge somewhere else before Ominimo gets there. The real lesson of Hungary is not “AI pricing wins” — it is that markets where incumbents rate on five or six variables, sold through aggregators with an annual switching season, are open to anyone who prices on a hundred. That description fits Turkish and Balkan motor liability, Southeast Asian compulsory motor, and Latin American auto markets far better than it fits Spain or the US, where Ominimo is headed anyway. A team copying the CEE-talent-arbitrage model — Olympiad-grade data scientists at regional pay, greenfield stack, local carrier fronting — can take the same 7%-in-a-year result in markets Ominimo has publicly deprioritised for its Western push. The second shift is line-of-business: the identical mispricing exists in CEE home and small-commercial insurance, where incumbent rating is even cruder and no switching-season insurtech operates. Ominimo won’t follow quickly: its roadmap, its Series B story and its Zurich relationship all point west toward the biggest, hardest markets, its licence application will consume management and capital for years, and a 130-person company chasing France and a 2027 US launch has no bandwidth for Bucharest-adjacent side quests beyond the two markets already announced.

Sources and further reading

• AI insurtech Ominimo bags its first investment at a $220M valuation (TechCrunch, April 10, 2025)
• Profitable Serbian-Hungarian InsurTech Ominimo hits unicorn status two years after launch, raises €20.1 million (EU-Startups, July 27, 2026)
• 1st Serbian unicorn! Ominimo hits $1.6B valuation after EBRD-supported Series B (The Recursive, July 2026)
• Ominimo hits $1.6bn valuation with profit-first model (FinTech Global, July 28, 2026)
• Hungarian insurtech Ominimo valued at $1.6 billion in EBRD-led round (The Insurer, July 28, 2026)
• Zurich takes a stake in AI-powered insurer Ominimo to drive European expansion (Insurance Business, April 2025)
• Serbian-Hungarian Ominimo hits $1.6B valuation, becoming the Balkans’ newest insurtech unicorn (Tech Funding News, July 2026)
• Ominimo Reviews (Trustpilot, accessed July 2026)
• SIGNAL IDUNA Biztosító Zrt. (Ominimo) biztosítónál érdemes kötni KGFB-t? (gyakorikerdesek.hu forum threads, 2025-2026)
• Motor Vehicle Insurance — Europe (Statista Market Forecast, 2025)