Guidewire to buy hyperexponential for $1.7 billion in cash to push agentic underwriting

Guidewire Software has spent two decades selling the core systems that property and casualty insurers use to run policies, claims and billing. On Tuesday the company moved to buy its way into the decision layer above those systems, announcing a definitive agreement to acquire London-based hyperexponential for a base price of $1.7 billion in cash, with up to $400 million more in contingent consideration tied to recurring revenue targets.

The announcement came at Guidewire's Connections conference in Las Vegas, where chief executive Mike Rosenbaum has been pushing a single argument: that the next phase of insurance software will be built around AI agents that work alongside underwriters rather than replacing them. Hyperexponential, known in the market as hx, sells AI-native underwriting and pricing software for commercial lines, a segment where risk is negotiated case by case and pricing judgement matters more than standardised rates.

Guidewire is not a startup. Its platform is used by more than 570 property and casualty insurers across 44 countries, and it has been selling into that base for years. Hyperexponential is a venture-backed company founded in 2017 that raised nearly $90 million from Andreessen Horowitz and Battery Ventures, according to PitchBook data cited by TechCrunch. The gap between those two profiles is much of the story of the deal.

The transaction values hx at roughly 19 times the equity it raised, a return that venture investors rarely see in enterprise software outside the current AI cycle. It also puts a mature incumbent in the position of buying the technology that investors have worried might hollow out its core market.

Key Facts

Guidewire said the base purchase price is $1.7 billion in cash, subject to adjustments, plus up to an additional $400 million if specified recurring revenue targets are met. Reuters reported on October 27 that the total potential value of the deal reaches $2.1 billion. The structure means the final price depends on how quickly hx's recurring revenue compounds inside Guidewire's sales channel rather than on the signing date.

The company will fund the purchase with $1 billion in new term loan financing plus existing cash on hand. Bank of America and Citibank have provided binding debt commitments. Guidewire expects the transaction to close in the second quarter of its fiscal 2027, the quarter ending January 31, 2027, and said it will not affect the first-quarter guidance it issued on September 3, 2026. Full-year guidance will be updated on the December 2026 earnings call.

On financial impact, Guidewire said the acquisition will add at least 650 basis points to fiscal 2027 annual recurring revenue growth on an inorganic basis, and at least 150 basis points to durable organic ARR growth starting in fiscal 2028. Those are company projections, not results, and they assume the close happens on the expected timetable.

Hyperexponential's software has powered underwriting decisions across more than $75 billion in annual commercial property and casualty premium for insurers, reinsurers and managing general agents. TechCrunch reported on October 27 that the company says nearly 600 insurers across 40 countries use its software, while Guidewire's own platform reaches more than 570 insurers in 44 countries. The overlap is substantial, which is precisely the point: the acquisition is partly a cross-sell motion into an installed base the buyer already knows.

Advisors on the transaction are Centerview Partners and Latham & Watkins for Guidewire, and J.P. Morgan and Wilson Sonsini Goodrich & Rosati UK for hyperexponential. Hyperexponential was already a Guidewire PartnerConnect Technology Alliance Partner before the deal was signed, so the two engineering and commercial teams have worked together in the past.

Analysis

Reinsurance News reported on October 27 that Guidewire frames the combination as a way for insurers to respond to more submissions faster, with pricing and risk decisions that reflect their best underwriting judgment. That is the commercial argument, and it is a coherent one. Commercial underwriting is a chain of judgement: a broker submits a risk, an underwriter assesses it, an actuary prices it, and the carrier decides how much of it to keep and how much to cede to reinsurers.

What this really means is that Guidewire is not buying capacity or distribution. It is buying the software layer where a human decision becomes a priced contract, and it is paying a premium because that layer is exactly where AI agents are most likely to be inserted. An agent that can read a submission, apply appetite rules, propose a price and document its reasoning is far more valuable sitting inside a core system that already holds the policy, the claims history and the rules than it is as a standalone tool sold alongside that system.

Insurance Innovation Reporter reported on October 28 that Rosenbaum called the deal maybe the most strategic thing he has done as chief executive. He argued that modern core applications already contain what AI agents need: the current state of a transaction, the rules governing what can happen next, and the history that gives a decision context. He drew a parallel to coding assistants, saying that just as engineers use Codex to code, underwriters will use agentic underwriting systems to underwrite. There is a discipline problem buried in that analogy. Coding assistants improve because code compiles or fails and tests pass or break, so the feedback loop is fast and cheap. Underwriting has no such loop, because a mispriced policy looks fine until losses emerge, sometimes years later.

The bigger picture here is consolidation of the AI layer by the system of record. Guidewire is paying up for a company that was already a partner, and the earnout structure suggests the buyer itself is unsure how fast the recurring revenue will compound. Tying $400 million to revenue targets rather than to a fixed close price shifts part of the integration risk onto hx's founders and backers, who now have to sell into Guidewire's base to collect the full amount.

Why It Matters

For insurers, the practical question is whether agentic underwriting changes loss ratios or simply speeds up paperwork. Hyperexponential chief executive Amrit Santhirasenan said the two companies will bring insight from submission to bind and from the individual risk to the entire portfolio, and that insurers will be able to respond to submissions faster and, more importantly, improve loss ratios and allocate capital more effectively. That second claim is the one worth watching, because faster submission handling is easy to demonstrate and better loss ratios are not.

For the market, the deal is a signal about where enterprise AI value is accruing. A London company that raised under $100 million in equity is being sold for roughly 19 times that amount, and the buyer is an incumbent rather than a private equity firm or a larger AI vendor. That is the reverse of the disruption narrative that has hung over core insurance software for years, and it suggests that owning the workflow and the data may matter more than owning the model.

For Guidewire's own investors, the near-term arithmetic is straightforward: $1 billion of new debt, a base price of $1.7 billion, and an ARR contribution the company has quantified in basis points rather than dollars. The stock reaction will depend less on the strategic story than on whether the December guidance update shows the acquired revenue landing on the schedule management has described.

Next Up

The immediate milestone is the close, expected in the quarter ending January 31, 2027. Guidewire President John Mullen said it was premature to describe the future roadmap combining hx with Guidewire's existing offerings, so the detailed product integration plan will not be public before then. Santhirasenan said hx began in pricing and rating but moved aggressively into AI after ChatGPT appeared, which is a hint about where the product emphasis may land.

Then comes the December 2026 earnings call, when management updates full-year guidance, followed by the fiscal 2027 numbers that will show whether the 650 basis point inorganic ARR lift and the 150 basis point organic contribution from fiscal 2028 are tracking. Santhirasenan's team will be selling into a base of more than 570 insurers in 44 countries, and the $400 million earnout depends on how many of them convert.