How Digitalization Saved Freedom Insurance $14 Million in One Year

In the past, our company was viewed mainly as a lead generator. That role is now being redefined. Over the last five years, we’ve grown eight- to tenfold and brought more than a million clients into the holding company’s ecosystem. Today we don’t just bring people in — we offer them a wide range of services through our SuperApp. Insurance isn’t just about selling a policy; it’s about resolving problems quickly after an accident. These services help us retain clients within the ecosystem and maximize their lifetime value (LTV).

We doubled our annual growth through 2025 and currently hold about 15% of the market. The 25% target still stands, but sustaining that pace is objectively harder now. Our growth rate slowed, as expected, in 2026. Further growth will have to come from new, unique products and from entering underserved market segments.

It’s the right call. For years, rates were set by law and didn’t keep pace with actual inflation. Since 2023, both the number and size of insurance payouts have risen sharply — from 100,000 to 167,000 per year. The regulator is shifting to a market-based approach, tying rates to actual loss ratios from prior periods. For half of Kazakhstan’s car owners, that means policy prices will rise more than 40%. This should bring the compulsory third-party liability (CTPL) insurance market back into balance and make the segment more attractive to insurers.

Our portfolio is diversified. Compulsory auto insurance makes up less than 20% of it, financial loss insurance accounts for 15%, and voluntary lines — medical, property and liability — each account for about 10%. We’re not dependent on compulsory insurance lines. We’ve built that resilience through multiple distribution channels: direct online sales via the SuperApp, partner integrations through our API, a traditional branch network and participation in the international reinsurance market.

We balance all of them. We have three core objectives: growing market share through sales, maintaining profitability so we can replenish capital in line with regulatory requirements and retaining clients within the ecosystem to increase LTV. The emphasis shifts from time to time, but over the medium term we develop all three in parallel.

This is the first market decline we’ve seen in five to 10 years. The main driver is CTPL, which has become highly unprofitable. In response, the market scaled back its sales channels, making policies harder to buy. Of Kazakhstan’s roughly 6 million registered vehicles, about 25 to 27% went without insurance. We’ve raised this with the regulator, and rates will be revised upward in July or August. We expect the market to stabilize by year-end.

In property insurance, natural disasters around the world caused major reinsurance risks assumed in Kazakhstan to materialize. That’s simply a normal part of doing business.

The CTPL market is extremely unprofitable. The payout-to-premium ratio reaches 170%. Our ceiling in the CTPL market is around 30% to 40%. Our DTP.kz platform helps us manage that burden. The investment has paid off 14 times over: in 2025 alone, we saved $14 million on claims by cutting out intermediaries. Our average claim payout runs 130,000 tenge lower than the industry average.

We no longer handle physical case files — 100% of claims go through DTP.kz. In early 2026, we hit an industry record of 5,000 payouts in a single month. Team efficiency has jumped dramatically: in 2024, one specialist processed 31 claims a month; now that figure is 87. Clients no longer need to bring us paper documents. We pull the necessary information directly from government databases. They inspect their own vehicles through the app, and our AI reviews the photos and video to assess the damage and automatically generate an appraisal report. About 70% of payouts are processed within 30 days.

We expect the new insurance law to pass by year-end. The infrastructure and government database integrations are already built. As soon as the legal framework is in place, we’ll launch a marketplace where clients can compare and choose products from different insurers on a single platform. Several insurers have already signaled they’re ready to participate.

The core issue is the economics of auto repair shops. Repair labor has thin margins, so shops rely on selling parts to make money. Kazakhstan doesn’t yet have repair-shop networks with integrated parts suppliers. We plan to address this through our Freedom Auto ecosystem project. Once that financial model is in place, customers will be able to have their cars repaired directly instead of receiving a cash payout.

We plan to launch cyber insurance by 2027. First, we need to consolidate the holding company’s expertise in data centers and cyber investigations. Our nearer-term focus is parametric insurance. We’ve already piloted flight-delay insurance for international flights and rolled it out for domestic flights in July.

The second area is smart agricultural insurance. Using satellite imagery, our AI models can identify which crop is planted in a given field with 95% accuracy, and the whole process is fully digital. This addresses one of farmers’ biggest problems — a lack of collateral — since the policy itself can serve as security for a bank loan.

Our back office already employs around 100 IT specialists, data scientists and AI engineers. We’re essentially more of an IT company than a traditional insurer at this point. In five years, our insurance license will still be the core platform through which we assume risk, but it will be wrapped in a much larger IT platform with API integrations and AI solutions.

As the holding company expands through its banking licenses, we’ll enter Tajikistan, Uzbekistan and Armenia with the DTP.kz platform and its AI-powered damage-assessment module. Our second product with strong export potential is agricultural insurance, which can be scaled relatively easily thanks to international reinsurance.

DTP.kz, by the way, isn’t a single-company platform anymore. Nine Kazakhstani insurers now use it to combat fraud.

By 2027, we want half of all claims under 1 million tenge to be approved entirely by AI, with no human involvement. The algorithm would independently assess the damage, price out the parts and offer the client a settlement, driving operating costs on those claims to zero.

It’s a hybrid. Our traditional agency network generates the margins we reinvest in development. At the same time, that traditional business is fully digitized — 100% electronic document management, direct integration with government databases. Our sales teams are twice as efficient as the market average, and our administrative costs run well below it.

A solid grounding in insurance still matters most. It’s a specialized, heavily regulated business. A startup mentality is great, but efforts to disrupt our processes can’t lead to chaos. We have an obligation to stay transparent and legible to external institutions.

Competence comes first. Someone can be difficult to work with, but if they’re truly skilled, we’ll find a way to make it work. Culture comes second. As the company grew, we brought together people from very different corporate backgrounds, so we launched the “Freedom DNA” project and defined three core values: leadership, innovation and accountability for results. Today, our employee Net Promoter Score (eNPS) consistently sits at 80%, and turnover is minimal.

The challenges. Insurance can look boring from the outside, and I want to change that perception. My own path here was a bit unusual. Most people move from sales into management, but I came up through analytics and actuarial work, and later worked on AI scoring models. That combination gives me a systems-level view and helps me come up with new ideas for the market.

The key is trusting the team. I don’t get involved in my senior managers’ day-to-day work. But I do stay close to clients. Every day, I read the complaints and thank-you messages that come in through the “Write to the Chairman” button on our website. If 15 complaints come in on a given day, that’s a signal that something in the system is broken, and it’s time to dig in.

It’s almost impossible to fully disconnect from work. For the first couple of years, I found myself thinking about work even on vacation. Now I just change my surroundings: I might work from the mountains or the coast for a while. Exercise helps too; regular physical activity lets my body recover and helps me refocus.

I love woodworking. I still use furniture at home that I built myself. Working with my hands is a great way to clear my head. I also love taking long walks through new cities when I travel. It gives me a fresh perspective.

Collecting data. Every morning I start with the websites of the National Bank, the regulator and the Swiss Re Institute rather than the news. I review the statistics and track trends. It’s an analyst’s habit, honed to the point where it’s become second nature.

As a CEO, it’s hard to write about everything; trade secrets and corporate ethics put real limits on that. I solved the problem with technology by building an AI agent called “AIzamat” and training it on 20 years of my own work and experience. It’ll be connected to my channel, so anyone can ask it about market dynamics or sales trends. The AI will collect statistics, generate charts and respond to users on its own. While I’m running the company, my digital twin will provide analytics.