More consumers favor insurers with modern platforms and are opting away from older offerings.

The startup landscape has never been more dynamic and rapidly evolving as it is today — especially in the insurtech market, which is projected to hit nearly $340 billion by 2032.

Despite this positive future outlook and overall opportunity, immediate-term economic hurdles are putting pressure on the insurance startup world. The venture capital market is undergoing a correction due to diminishing expectations of Fed rate cuts (though quite likely in 2025 with the latest inflation data) and the high cost of public and private lending for investment funds.

This has led to increased scrutiny of investment opportunities, with a focus on metrics such as execution excellence, customer experience, churn, unit economics, and runway to profitability. As a result, funding rounds have become less frequent, and founders are actively seeking ways to reduce costs and extend their financial runway.

Despite these economic headwinds, the startup community, particularly the insurance sector has attracted record investments driven by SaaS products, GenAI, and advanced analytics offerings.

These forward-focused startups are reshaping insurance, underwriting, and claims, challenging traditional players. Innovative metrics now enable real-time risk assessment, making consumers more comfortable with their premiums. Consumers increasingly favor insurers with modern platforms and opting away from "the way it used to be" offerings.

This shift pressures startups to balance profitability with innovation, with technology playing a central role in cost reduction strategies. Common themes emerge from discussions with startup founders across various stages of maturity:

Automating net new customer onboarding experiences

Direct-to-consumer (D2C) insurance players are accustomed to dealing with document-heavy customer onboarding. While these documents are fairly digitized, they are far away from nationwide standardization.

Traditional players still have significant human intervention throughout their processes. Insurtech is differentiating by leveraging AI to pull relevant customer information across multiple non-standard documentations usually submitted. Even during a claims process, AI is being heavily leveraged to automate incident analysis, generate incident reports, and provide quick case updates to the affected individual.

Furthermore, GenAI insurance agents are helping customers make the right choices with their insurance tiers, submit accurate documents, and provide live feedback with very subtle intervention from humans. Such asset-light operations are creating significant savings opportunities and often, an improved experience.

In addition, these D2C players are taking it a step further by developing robust customer data platforms (CDPs) and continuously integrating first- and third-party data related to their customers. A well-designed CDP can significantly reduce customer churn and unlock upsell/cross-sell opportunities.

DevOps investment optimization by B2B SaaS insurtech startups

A few very niche startups are actively helping larger insurance firms with very specific processes, such as risk assessments, underwriting, securitization of premiums, claims management, payments, etc.

For most early-stage startups, having dedicated DevOps engineers doesn't make financial sense. Managing tens of customers is still feasible, but late-stage startups managing hundreds, if not thousands, of customers, spend their precious cash to hire more dedicated DevOps engineers.

This is where mature startups work towards architecture frameworks on the public cloud that automate tenancy onboarding in single and multi-tenant environments. This approach is a game-changer in terms of efficiency, with some founders reporting a tenfold reduction in engineering operating expenses.

AI Hardware for AI Startups

There won't be a single startup board remaining that hasn't discussed their GenAI strategy in the last 18 months. More than $500bn have been invested into GenAI startups, compared to $100bn+ in the next biggest investment segment (EV Charging Startups) in the last 12 months globally.

Whether one is a creator or consumer of LLMs — Gemini, Anthropic, OpenAI, Llama, et al have become the talk of town. AI Infrastructure supporting this rapidly growing market continues to be impacted by supply-demand disparity. Some niche startups are developing LLMs specifically trained for insurance providers and the broader insurance value chain.

This involves extensive training of these AI models with insurance-specific datasets. Model training is going to be an 'ALWAYS ON' activity for such startups, and like everyone else, they need AI hardware. While public cloud platforms are creatively assisting startups in gaining timely access to GPUs, TPUs are gaining a fair bit of interest.

In certain cases, TPUs reduce ML model training time and cost by 25 to 35% compared to GPUs. Startups are seeking expert advice to refactor models for TPU compatibility and maintain diversification in the silicon shortage market.

Optimizing internal processes

Startups are leading the charge when it comes to adopting GenAI to enhance employee and developer productivity. This includes developing internal code-assist agents to streamline coding and automating code testing to increase efficiency and accuracy. These advancements yield substantial operational benefits, both tangible and intangible. Additionally, GenAI is being applied to various routine HR, finance, and marketing processes.

Data enablers

It should be far from a surprise that the insurance industry not only has massive historical datasets, but they are collecting more and more data every day at an exponential rate. Imagine the amount of data those mile-wise vehicle insurance plans must be collecting when they are tracking your driving style every second.

These platforms are the epitome of query performance, lightning-fast analytics, and accessibility to data on demand. Hyperscalers have made considerable investments in building 'managed' solutions for ETL pipelines, serverless DataWarehouse platforms that enable native ML/Analytical capabilities, and BI dashboard toolkits.

Moreover, products dedicated to data engineering such as Databricks and Snowflake are giving more niche data lake/data warehouse capabilities to govern this data. However, we are seeing more and more startups throwing money at computing behind EDW platforms to purchase performance in the early stages of product development, resulting in poor price and performance. This approach may be a shortcut to win customers early on, but has resulted in sub-par query write-ups, that in some cases continue to cascade further as the product grows.

Eventually, in search of better prices and performance, they have spent more cash switching between top providers (hopping between known players and platforms), but the dollars continue to evaporate. Teams that invested early in building internal expertise or sought external help to optimize queries, business logic, and algorithms have achieved greater long-term success.

The current landscape presents significant challenges for insurance startups seeking to show value to investors and customers. Founders, already facing a demanding and lonely role, must navigate these choppy waters with heightened awareness and precision.

By focusing on key areas such as operational efficiency, strategic partnerships, and technological innovation, startups can weather the storm and position themselves for accelerated growth and profitability.

Deepankar Mathur has been with Searce for eight years and leads Searce's North America Consulting business dedicated towards high growth venture backed Startups in the region. In the last several years, he has strategically advised over 500 startups across technology initiatives, go-to-market planning and business models to accelerate the path to revenue & profitability.

Originally published in Property Casualty 360