• AI visibility

Guide · AI search

How can an insurtech win customers from incumbent insurers when shoppers ask AI?

By becoming the alternative an assistant can explain and vouch for: a clear category story, honest comparisons with the big carriers, and reviews and coverage that answer “is this legit?”. AI already sits in the insurance shopping journey, and shoppers who use it switch more often, which favors challengers. But assistants default to familiar brands, so an insurtech has to earn its place in the answer.

By Underneath Team Guide Updated

The short version

  1. AI is becoming a customer source for insurtechs: Insurify (opens in a new tab) says the share of its new customers who found it through ChatGPT and AI conversations rose from 0.15% in the first half of 2025 to 5.7% in the first half of 2026.
  2. AI users are switchers: J.D. Power (opens in a new tab) found 32% of auto insurance shoppers used AI tools, and they were more than 1.3 times as likely to switch insurers.
  3. Customers are expensive to win: Lemonade (opens in a new tab) spent $78 million on sales and marketing in the second quarter of 2026, a quarter in which it added about 166 thousand customers.
  4. Distribution is shifting to partners: Root (opens in a new tab) said partnership and independent agent channels produced about 51% of its new writings, up from about 44% a year earlier.
  5. Early-stage money is tighter: Gallagher Re (opens in a new tab) reports early-stage insurtech funding fell 51.8% in the second quarter of 2026, to $264.19 million.

A note before you read: this article is about how insurtech brands appear in AI answers. It does not recommend any policy, and none of it should be read as legal or regulatory guidance for a licensed carrier or agency.

Why is AI search a different problem for an insurtech than for an incumbent?

Assistants lean on well-documented brands, and incumbents have decades of documentation while a startup has a few years.

State Farm, Geico and Progressive have years of ratings, press, comparison-site listings and customer reviews behind them. An insurtech may have a better product and a fraction of that record. Two studies show what that means in AI answers:

  • Known by name, missing from discovery. In a study of 112 Product Hunt startups (opens in a new tab) by Amit Prakash Sharma, ChatGPT recognized the products 99.4% of the time when asked about them by name, but named them in only 3.32% of open discovery questions.
  • Familiar brands win ties. In a controlled test (opens in a new tab) by Xi Chu and YuPeng Hou, assistants recommended the well-known brand 100% of the time when products had identical specifications. A competitor’s rating edge of less than 0.1 stars was enough to break that.

Neither study covered insurance. Our inference for insurtechs: when a shopper asks an open question, such as “who has the cheapest renters insurance?”, the assistant will reach for familiar carriers unless it finds specific, credible evidence that a newer option is different. The incumbents’ side of this contest is covered in how carriers and agencies win quotes from AI.

The funding climate raises the stakes. Gallagher Re counted $2.44 billion of insurtech funding in the second quarter of 2026, the most since 2022, but 99.1% went to AI-focused companies and early-stage funding fell sharply. For most startups, we infer, a channel built on evidence rather than ad spend is worth more when capital is concentrated.

Where are insurtechs already meeting shoppers inside AI assistants?

Inside the assistants themselves, through in-chat apps, and in the answers shoppers read before they request a quote.

Insurify, a digital insurance agent, shows the in-chat route. In February 2026 it launched what FinTech Global (opens in a new tab) describes as the industry’s first insurance app in ChatGPT, drawing on more than 196m auto insurance quotes and over 70,000 verified customer reviews. In August it added real-time quotes (opens in a new tab) for shoppers in a first group of states, including Arizona, Ohio and Wisconsin. Insurify reports that in the six months after launch, ChatGPT-referred traffic grew 151% and revenue from those visitors grew 212%. These are the company’s own figures, but they are the clearest public numbers we found on AI as an insurance acquisition channel.

The broader shift shows up in J.D. Power’s data. Shoppers who used AI were more than 1.3 times as likely to switch insurers. J.D. Power reads this as customers turning to AI when insurers fail to explain coverage, which moves control of information away from carriers.

That matters more to a challenger than to an incumbent. Our inference: a shopper who is willing to switch, and is asking an assistant to explain the options, is exactly the shopper an insurtech needs to reach, provided the assistant knows the insurtech exists.

Which questions decide whether an insurtech is considered?

Alternative, comparison and legitimacy questions, plus questions about new kinds of coverage. The examples below are our own drafts of typical questions, not prompts captured from real shoppers or partners.

StageIllustrative prompt
Alternatives“Alternatives to Geico for a new driver with a clean record”
Comparison“Lemonade vs State Farm for renters insurance in Illinois”
Category“Is pay-per-mile car insurance worth it if I work from home?”
Legitimacy“Is Root insurance legit, and do they actually pay claims?”
Embedded“Should I buy the insurance my car dealer offers at checkout?”
Partner search“Embedded insurance providers for an online used-car marketplace”

The legitimacy question deserves special attention. In our study of “Is this brand legit?” answers across 79 brands, every complete answer said the brand was legitimate, but 99.7% raised at least one problem. 88.0% of answers cited a review or complaint platform, and Trustpilot and the BBB made up 61.7% of review-platform citations. For an insurtech, how complaints are handled on those platforms becomes part of the answer.

How does AI visibility turn into customers for an insurtech?

It puts the insurtech on the shortlist of a shopper ready to switch, who can quote and buy in minutes.

For a direct insurtech, the path is short. A shopper asks about alternatives, the answer names the company, the shopper checks reviews, opens the app, gets a quote and binds. The economics show why that path is valuable. Lemonade ended the second quarter of 2026 with 3,308,666 customers, up 23%, and in-force premium of $1.43 billion, up 32%. Premium per customer was $433. It also spent $78 million on sales and marketing in the quarter, up from $60 million a year earlier.

Lemonade has grown from 1.00 million customers at the end of 2021, according to The Motley Fool (opens in a new tab), mostly by paying to acquire them. Our inference: every customer who arrives because an assistant named and explained the company is one the marketing budget did not have to buy. Insurify’s reported rise in AI-sourced customers, from 0.15% to 5.7% of new customers in a year, shows how fast that share can move once a company is present where shoppers ask.

How does embedded insurance change the AI question?

It adds partners choosing an insurance provider, and shoppers deciding whether to trust the policy offered at checkout.

Distribution is moving toward partners. Root, which now operates in 37 states with 483,921 policies in force, said partnership and independent agent channels made up about 51% of new writings, while its direct channel “remained challenging.” Lemonade’s car product is now available in states representing nearly 50% of the US car insurance market, and offers Tesla owners 50% off every mile driven using Tesla’s Full Self-Driving technology.

Shoppers are open to it. J.D. Power’s 2026 shopping study (opens in a new tab) found 36% of recent shoppers interested in buying embedded insurance sold through a car dealer or manufacturer. Usage-based insurance, a common insurtech product, was used by 34% of those buying from a new insurer.

Our inference for embedded providers: two kinds of AI question matter. Shoppers ask whether the policy offered at checkout is a good deal, so the provider needs a clear, independent reputation. And partners, such as marketplaces, dealers and lenders, research providers too, so business-facing pages and trade coverage matter as much as consumer reviews.

What decides whether an insurtech is named?

Independent evidence of a real difference, a visible review record, and clear facts about what is sold where. Platforms document only part of this.

Documented by platforms. Google says its systems give even more weight to strong expertise and trust signals on “Your Money or Your Life” topics (opens in a new tab), which include financial stability.

Observed in studies.

  • Independent coverage. In our brand study, each tenfold increase in independent sites naming a brand went with 4.7 times the odds of being recommended.
  • Search foundations. In the Product Hunt study, referring domains, an ordinary search signal, predicted which startups Perplexity named.
  • A visible difference. In the controlled brand test, a small, verifiable advantage broke the default toward the known brand.
  • Review platforms. In our legitimacy study, review and complaint platforms carried most of the negative claims assistants made.

Our inference for insurtechs. The difference has to be specific and checkable: a price structure that is genuinely different, a claims process that reviewers confirm, a coverage feature incumbents lack, and plain facts about which states you write in. Inflated claims are a risk in a regulated industry, and GEO can backfire when it outruns the facts.

What does GEO look like for an insurtech?

For an insurtech, generative engine optimization (GEO) means building, quickly and precisely, the evidence incumbents already have.

  1. Own the category explainer. Explain usage-based, pay-per-mile or embedded insurance honestly: how it works, what it costs, who it suits and who it does not.
  2. Compare yourself with incumbents honestly. Keep comparison pages current on price structure, coverage and claims. Our research on whether comparison pages help brands get cited shows what tends to work.
  3. Answer the legitimacy question in public. Maintain profiles on Trustpilot, the BBB and app stores, and respond to complaints, since these are what assistants cite.
  4. State availability plainly. Which products you sell in which states, kept up to date as you expand.
  5. Earn independent coverage. Insurance trade press, personal finance media and comparison platforms give assistants third-party evidence.
  6. Serve partners too. Embedded providers need clear business-facing pages and trade coverage for the platforms evaluating them.
  7. Check launches. New products are often missing at first; why ChatGPT misses newly launched products explains why.

The broader playbook for challengers is in how a small brand can get recommended by AI assistants, and the wider evidence on incumbents is in do AI assistants favor big brands?. No one can guarantee an assistant will recommend a startup; GEO makes the evidence it finds accurate and easy to verify.

Where is the evidence on insurtechs still thin?

On how assistants treat insurance startups specifically, and on how many policies AI answers produce across the industry.

  • No insurance-specific test. The studies of new and incumbent brands used startups across categories and consumer products, not insurers.
  • Company-reported figures. Insurify’s growth numbers are its own, from a company with a product in ChatGPT, and not independently audited.
  • Surveys. J.D. Power’s AI figures describe what shoppers report, not what assistants told them.
  • No industry-wide link to policies. Beyond one company’s figures, we found no public data connecting insurtech AI visibility with policies written.

Where should an insurtech start?

Ask assistants the alternative, comparison and legitimacy questions your target customers ask, and note which incumbents and sources win.

That first check shows whether assistants understand your category, whether they name you next to the carriers you compete with, how they answer “is it legit?”, and whether your state availability and pricing model are described correctly. It also shows which reviews, articles or comparison sites the incumbents have that you lack.

If your growth plan depends on lowering what you pay to win each customer, or on winning partners for embedded distribution, ask us to map your AI visibility against the incumbents. We will test the questions that decide whether you are considered, trace the evidence behind each answer, and plan the coverage, content and review work, checked with your compliance team, that gives you a fair chance of being named. To see how a challenger’s category explainers, comparison pages and review profiles are planned and then tracked against incumbents, read our generative engine optimization service page.

Frequently asked questions

Can a new insurer appear in ChatGPT answers at all?

Yes, but rarely at first. In one study, startups were recognized 99.4% of the time when named, but appeared in only 3.32% of open discovery questions.

Do in-chat apps replace being named in AI answers?

No. An app helps shoppers who already chose it, while answers decide who gets considered. Insurify uses both: an app in ChatGPT and a brand that shoppers discover through AI conversations.

How do bad reviews affect an insurtech in AI answers?

They tend to appear. In our legitimacy study, 99.7% of answers raised at least one problem, often drawn from review and complaint platforms, so visible, well-handled complaints matter.

Should embedded insurance providers care about AI search?

Yes. Shoppers ask whether a checkout policy is a good deal, and partner platforms research providers before signing, so both consumer reputation and business-facing evidence matter.

Sources

Free strategy call

Some questions are easier to answer about your own business.

Bring the one that matters most. On a free 30-minute call we’ll take a first look at it and send you a short written read afterward.

Ask about my businessHow we work