The company told Fortune it surpassed $200 million in ARR in June โ a floor, not a point โ and that figure makes this round’s arithmetic unusually checkable.
Every EliseAI figure here is a company claim made to Fortune. There is no filing, no audit and no independent verification.“Surpassed $200 million in annual recurring revenue in June” is a floor rather than a point, and it is a June figure. The multiple of about 20 times and the valuation step of about 1.8 times are this publication’s arithmetic, and the 20 times is an upper bound.
Fortune assigns this round no Series letter; the prior round is the Series E. Nothing below offers a view on whether either valuation is justified. Nothing here is investment advice.
Fortune reports that EliseAI has raised $350 million at a $4 billion valuation, with the round co-led by existing investors Andreessen Horowitz and Bessemer Venture Partners alongside new investor Ontario Teachers’ Pension Plan, and participation from Sapphire Ventures and Navitas Capital. Fortune assigns no series letter to this round; the prior round was a Series E. The company told Fortune the round consisted entirely of primary capital โ no secondary sale for early investors or employees โ and that Sameer Dholakia of Bessemer is joining the board. It is the fourth time EliseAI has raised from the two lead venture firms since 2023.
The sentence that makes this round worth analyzing is the revenue one. EliseAI said it surpassed $200 million in annual recurring revenue in June, marking its fifth consecutive year in which revenue doubled year over year. Every figure attributed to the company here is a company claim made to Fortune โ in the article’s own words, “the company told Fortune,” “the company says,” and “EliseAI said.” There is no filing, no audit, and no independent verification.
On reach, the company told Fortune the platform powers roughly one in six apartments in the United States, with more than 30 million Americans having interacted with it, and that it handles roughly 5 million calls a month across its housing and healthcare businesses combined. EliseAI is headquartered in a 109,000-square-foot former Tiffany & Co. building in Manhattan and plans a second engineering hub in San Francisco.
Funding Trajectory โ Company Claims
2023 and prior: Multiple rounds from Andreessen Horowitz and Bessemer โ this is the fourth raise from both firms since 2023.
~August 2025 โ Series E: $250 million Series E at a $2.2 billion valuation โ the prior named round on record.
June 2026: Company told Fortune ARR surpassed $200 million โ a floor, not a ceiling, and a figure now three months old.
September 29, 2026: $350 million raise at $4 billion valuation announced to Fortune. No series letter assigned.
The key insight: “Surpassed $200 million” is a floor, not a point estimate, and it is a June figure โ meaning it is three months old at the time of announcement. The valuation arithmetic is therefore an upper bound: this publication’s calculation of roughly 20 times ARR uses the smallest defensible denominator. The actual multiple, against a current ARR figure, would be lower. That is not a judgment on the valuation; it is a statement about what the disclosure makes checkable.
Every EliseAI figure here is a company claim made to Fortune, with no filing behind it. The Island column comes from a release this publication read in full.
The Structural Read
Five days before this announcement, this publication read Island’s release disclosing a $400 million raise at a $6.4 billion valuation. Island’s only revenue statement was that it “has doubled annual recurring revenue every fiscal year since its 2022 launch” โ a rate with no base. No absolute ARR figure, no revenue number, and no customer count appeared anywhere in that document. The same valuation arithmetic that applies to EliseAI is simply not available for Island: there is no denominator.
That is the whole contrast, and it is worth naming precisely. Two rounds in the same week, in the same broad category of enterprise AI software. The one carrying the smaller valuation is the one that named the number its price rests on. EliseAI at $4 billion disclosed a floor ARR figure and a growth rate. Island at $6.4 billion disclosed only the rate. A rate without a level cannot be priced โ not because the level doesn’t exist, but because it wasn’t in the document.
This publication passes no judgment on either valuation and has not built a comparable set. No comparable set exists in either document. The observation is only about what each announcement makes checkable.
On the step between rounds: the prior round was a $250 million Series E at approximately $2.2 billion, roughly 13 months ago. The current valuation of $4 billion is roughly 1.8 times that figure โ this publication’s arithmetic, applied to company-stated numbers.
Structural Principle
“A smaller valuation that discloses more gives the market more to work with than a larger valuation that discloses less. Disclosure and price are separate variables. The discipline worth tracking is which founders treat them that way.”
One more structural distinction is worth drawing cleanly. This publication has written five pieces this week about vendors building permission infrastructure โ a capped agent wallet, credentials as a first-class component, an acquisition of robotic process automation, an approval gate on publish, send and spend, and an enterprisecontrol plane. EliseAI is not a sixth entry in that cluster, and saying so matters more than the pattern does.
CEO Minna Song told Fortune that Apollo, EliseAI’s agent, “takes actions within the permissions that it’s given,” that binding decisions โ approving or denying an application, sending a formal notice, signing off on a lease term โ “still go through a person,” and that on fair-housing queries Apollo drafts a response, flags the policy, and queues it, “but it doesn’t get the final say.” What Song describes is human-in-the-loop productdesign inside a regulated workflow.
That is ordinary and long-standing practice in fair-housing and healthcare software. It is not a separate permission layer sold as infrastructure to other builders. The difference between a product that asks permission and a product that sells permission is the distinction worth keeping. EliseAI is the former.
Three Implications
DISCLOSURE IS A PRICING INPUT
When a company states a floor ARR figure and a growth rate, external observers can construct an approximate revenue multiple and stress-test it against the valuation. When a company states only a growth rate, that arithmetic is impossible regardless of how strong the underlying business is. Founders and investors choosing what to disclose in a funding announcement are making a decision about how much external pricing discipline they invite.
The EliseAI announcement invites more of it than Island’s did โ not because EliseAI intended a lesson, but because the numbers were there.
VERTICAL AI COMPRESSES DEAL CYCLES, NOT JUST COSTS
Five consecutive years of doubling revenue, in a sector โ residential property management โ that is not known for fast procurement cycles, suggests that EliseAI’s value proposition lands at the operational level rather than the strategic one. Leasing inquiries, maintenance calls, and application processing have clear throughput metrics that a landlord or property manager can read against a monthly invoice. Vertical AI that attaches to a measurable workflow has a shorter path to a signed contract than horizontal AI that requires a customer to define its own use case.
The housing footprint the company described to Fortune โ roughly one in six U.S. apartments โ is consistent with that dynamic, though it is a company claim and cannot be independently verified here.
REPEAT LEAD INVESTORS CARRY A SIGNAL, BUT ONE WITH LIMITS
Andreessen Horowitz and Bessemer co-leading for the fourth time since 2023 is not evidence that the business is strong โ insiders have information asymmetry in both directions โ but it does mean that the two firms with the most direct access to EliseAI’s actual metrics chose to re-up rather than pass. Ontario Teachers’ Pension Plan joining as a new investor at this stage is a different signal: large institutional investors at late-stage rounds are typically operating from a different return profile and time horizon than early venture, which changes the composition of the cap table in ways the announcement does not fully describe. No headcount, margin, burn rate, or housing-versus-healthcare revenue split appears in the announcement, and none is supplied here.
Business Engineer Framework
The Map of AI โ Where Vertical SaaS Sits in the Stack
EliseAI occupies a specific layer in the AI stack: a vertical application built on top of foundation models, distribution infrastructure, and workflow APIs, sold into a regulated industry with defined compliance requirements. The Map of AI traces all nine layers of that stack and identifies where durable margin tends to concentrate โ and where it tends to erode. Understanding which layer EliseAI actually defends is the question the $4 billion valuation implicitly asks.
EliseAI disclosed a floor ARR figure and a five-year growth rate; this publication’s arithmetic puts the valuation at roughly 20 times that floor โ an upper bound, because “surpassed” is a minimum and June is three months back โ and the step from the prior round is roughly 1.8 times on the same basis; every one of those figures is a company claim made to Fortune, with no filing or audit behind it, and nothing here predicts what comes next; the structural observation that matters is simpler than the numbers: in the same week, the smaller round named the denominator and the larger one did not, and that asymmetry is entirely about what each announcement makes checkable, not about which business is stronger.
Source: Fortune โ EliseAI raises $350 million at $4 billion valuation. All EliseAI figures are company claims made to Fortune; no filing, audit, or independent verification exists. The ~20x ARR multiple and ~1.8x valuation step are this publication’s arithmetic and upper bounds. Nothing here is investment advice and nothing here predicts any outcome.
Every EliseAI figure above is a company claim made to Fortune, in the article’s own words “the company told Fortune”, “the company says” and “EliseAI said”. There is no filing, no audit and no independent verification behind any of them.“Surpassed $200 million in annual recurring revenue in June” states a floor rather than a point, and it is a June figure reported at the end of September.
The multiple of approximately 20 times and the valuation step of approximately 1.8 times are this publication’s arithmetic; because the revenue figure is a floor and three months old, the 20 times is an upper bound rather than an estimate. Fortune assigns this round no Series letter anywhere. The prior round is the one named a Series E, and no letter has been supplied above.
Nothing above offers a view on whether either the $4 billion or the $6.4 billion valuation is justified. Neither document contains comparable transactions and this publication has not assembled a comparable set. The Island comparison rests on that company’s release of 24 September 2026, which this publication read in full and reported on; its only revenue statement is a growth rate with no base. No headcount, customer count as a number, margin, burn or housing-versus-healthcare split appears in what was read, and none is supplied above.
On a public listing Minna Song said she did not think the company had any specific timelines or definitive outcomes, and nothing is extrapolated from that. Nothing above predicts anything about EliseAI, Island or either valuation, and nothing here is investment advice.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
Scroll to Top
Discover more from FourWeekMBA
Subscribe now to keep reading and get access to the full archive.
