Notch lands $30 million Series A to automate customer experience with AI agents - CTech
Notch was founded in 2021 to insure Instagram accounts against hacks. To do that, it needed every policy decision to be auditable and defensible to a regulator. That is an expensive problem to solve manually. So Notch built AI tooling to solve it — and discovered that every other insurer and financial institution has the same problem. The internal tool became the product. That origin story — not a pivot from one sector to another, but a tool built from necessity that found a market — is what Notch's $30 million Series A is designed to fund.
The round was led by Headline, with participation from Lightspeed Venture Partners, Jibe Ventures, Illuminate Financial, and Phoenix Insurance, bringing total funding to $45 million, according to a press release published by the Manila Times. Rafael Broshi, Notch's co-founder and chief executive, frames the compliance infrastructure as a structural moat. "Every decision in insurance has to be audit-ready," he said in the announcement. "That is a structural advantage — the audit trail is not a feature we bolted on, it is how the product was designed from the start." The company sells what it calls an "AI operating system for regulated industries" — AI agents that handle broker and policyholder interactions on the conversational side and claims processing and underwriting on the back-office side, with five layers of guardrails covering conversation safety, prompt injection defense, access rules, business limits, and jurisdiction rules.
The traction numbers are Notch's own. The company says annual recurring revenue grew 12x over the past 12 months, a claim that has not been independently audited. It reports having processed 10 million support tickets end-to-end, with an 87 percent resolution rate, a figure corroborated in a Microsoft Start-ups blog post describing Notch's Azure-based deployment. Those two numbers — the growth multiple and the resolution rate — travel together in Notch's announcement materials, and readers should know they come from the same unverified source.
The 87 percent number also shows up in a Guardio case study: the cybersecurity firm cleared a 20,000-ticket backlog in less than a week using Notch's AI agent and now resolves 87 percent of incoming tickets without human intervention, according to a PR Newswire release citing Amos Peled, Guardio's chief executive. The Guardio deployment is the most specific claim in Notch's public record, and the CEO quote gives it independent corroboration — but the 87 percent baseline and the resolution standard deserve scrutiny. Notch does not publish what share of those resolved tickets required human review before closure, which is the difference between a meaningful automation rate and a headline number.
There is a harder verification gap worth flagging before these numbers travel beyond this article. Notch says it holds SOC 2 Type II and ISO 27001 certifications. Both claims appear only on the company's own website. Neither has been independently confirmed via an audited report or third-party audit registry. For regulated-industry customers evaluating a vendor whose entire pitch rests on compliance credibility, that is not a minor caveat.
Elool Jacoby, co-founder and chief product officer, pointed to the Microsoft Azure deployment and those certifications as "the trust substrate that regulated customers require." That framing is accurate as a description of what regulated customers want. Whether Notch's certifications meet that bar independently is a separate question — one that security and compliance teams at prospective customers will answer for themselves.
The origin-as-necessity narrative is compelling, but it deserves scrutiny. Notch was founded as a specialty insurer in 2021. A $15 million seed round in September 2025 positioned the company as a customer experience platform for SaaS companies — e-commerce, gaming, the usual suspect verticals. The Series A framing is substantially different: insurers, financial institutions, and regulated industries broadly. The progression from insurer to SaaS CX to regulated-industry AI OS is not a straight line, and "we always intended to build this" is a convenient way to paper over a reframe. The compliance-first origin does have a logical connection to the current product — audit trails and traceability are genuinely what regulated industries require. Whether the narrative was constructed retroactively or reflects a genuine product evolution is worth independent verification.
One structural note on the cap table: Munich Re Ventures and LionTree were seed investors in that 2025 round. Neither appears in the Series A investor list. Changed investor composition does not necessarily signal anything operational — investors pass on rounds for all sorts of reasons. But the shift is worth noting given how heavily the announcement leans on the origin story.
The McKinsey estimate that generative AI could add $50 billion to $70 billion in incremental revenue to the insurance industry gives the market sizing context, though it is a forward-looking projection rather than a current measurement of market size. If the compliance requirements that make insurance hard also make it a defensible niche, the addressable market is real. The Microsoft Marketplace listing gives Notch a distribution path into enterprise accounts without a full direct sales buildout.
What is worth watching: whether the 12x ARR growth is a genuine product-market fit signal or a base effect from a small starting number. Seed-stage companies hitting growth milestones in low six figures can claim extraordinary multiples without it meaning much. And whether the story Notch is telling today — the insurer that had to build compliant AI and turned it into a product — is the same story it will be telling 18 months from now is a separate question that the $30 million will help answer.