A real estate feasibility study is a document that costs a fortune and takes forever, which is exactly why anyone building a new shopping center or apartment block needs one. It is the financial and market due diligence that convinces a bank to lend tens of millions. For decades, producing one meant hiring a boutique consultancy, paying a six-figure fee, and waiting three months. Feasibly, a new startup from Park City, thinks it can do the same job for ten thousand dollars and in under a week [Brief Glance, December 2025]. The bet is not just on speed, but on whether an AI, guided by human experts, can produce analysis a regulated financial institution will trust with its capital.
The wedge is time and money
Feasibly’s platform uses what it calls a multi-agent AI system to ingest project data and publicly available information on demographics, local competition, and construction costs. It then generates a narrative report covering financial modeling and cash-flow projections, all styled as “bank-ready” [Feasibly website, December 2025]. The company claims this process, which includes expert human oversight, takes three to five days [Feasibly how-it-works, 2026]. For a developer racing against option expirations or rate hikes, that compression of calendar risk could be transformative, even before counting the cost savings. The pricing starts at $10,000 for a single study, with tiered plans for multiple reports or a subscription, positioning it as an accessible tool for smaller developers who might have skipped a formal study altogether [Feasibly website, December 2025].
A team built on decades of analysis
The founder, Brian Connolly, is not a software engineer pitching a tech solution in search of a problem. His background is in the very work Feasibly aims to automate. Public records show over two decades in financial feasibility assessment and economic impact analysis, specifically for large sports and entertainment facilities [Columbia University SPS, 2026]. His strategic hire, Russell Scibetti, brings experience from the front office of the New York Football Giants, a world where multi-million dollar decisions hinge on detailed projections [LinkedIn, 2026]. This grounding in the old way of doing things is their first line of defense against the biggest risk: that an AI-generated report will be seen as a cheap approximation, not a serious piece of analysis.
| Role | Name | Key Background |
|---|---|---|
| Founder & CEO | Brian Connolly | 20+ years in financial feasibility and economic impact analysis [Columbia University SPS, 2026]. |
| SVP, Strategy & Business Intelligence | Russell Scibetti | Former SVP at the New York Football Giants [LinkedIn, 2026]. |
Where the model meets the market
The competitive field is nascent but active. Tools like Zenerate and GrowthGrid also apply AI to real estate analytics, often focusing on site selection or preliminary pro formas. Feasibly’s distinction is its direct aim at the final, authoritative document that seals a deal. Its success hinges on a few critical, unproven assumptions. The “expert human oversight” is vague; is it a light review or a deep, line-by-line validation? At a $10,000 price point, the economics of employing seasoned analysts to scrub every report are tight. Then there is the adoption curve. A mid-sized regional bank’s credit committee is a conservative audience. They will need to see a track record of Feasibly reports that accurately predicted project outcomes before they accept them as a substitute for a known consultancy.
- The accuracy covenant. The product’s utility vanishes if its financial projections are off by a few percentage points. The company’s reputation, and its clients’ projects, live or die on the precision of its AI’s assumptions about rent growth, occupancy, and construction delays.
- The human-in-the-loop cost. The promised expert review is the crucial quality gate, but it is also the main cost center. Scaling this service without degrading quality or exploding prices will be a fundamental operations challenge.
- The incumbent inertia. The major consulting firms Feasibly seeks to displace are not standing still. They have deep client relationships and are undoubtedly building their own AI-assisted tools, which they can offer as a premium upgrade to their existing service.
The next twelve months
With a $1 million pre-seed round closed at its launch in late 2025, Feasibly’s immediate task is to prove its model works in the wild [Brief Glance, December 2025]. The key metrics to watch will be customer names and renewal rates. A handful of named developer clients using the service for actual bank financing would be a powerful signal. So would a transition from one-off reports to multi-study subscriptions, indicating the product has become a trusted part of a developer’s toolkit. The company will also need to clearly articulate the division of labor between its AI and its human experts, transforming a marketing claim into a credible quality assurance process.
On paper, the unit economics are intriguing. If a traditional feasibility study costs $100,000 and Feasibly can deliver a comparable product for $10,000, the value capture is immense. The math only works, however, if the “comparable” part holds. A back-of-the-envelope calculation: to replace a single $100k consultant study, Feasibly needs to sell ten of its $10k reports. But if it takes one expert analyst a week to properly vet each AI-generated report, the salary burden quickly eats the margin. The company it must beat is not another AI startup, but the entrenched, person-to-person trust of a firm like HR&A Advisors or a Big Four consultancy. Feasibly’s AI doesn’t need to be perfect. It just needs to be good enough for the banker to sign the loan.
Sources
- [Brief Glance, December 2025] Feasibly's AI Cuts Real Estate Analysis from Months to Days | https://briefglance.com/articles/feasiblys-ai-cuts-real-estate-analysis-from-months-to-days
- [Feasibly website, December 2025] Feasibly product and pricing information | https://www.planfeasibly.com
- [Feasibly how-it-works, 2026] Feasibly process description | https://www.planfeasibly.ai/how-it-works
- [Columbia University SPS, 2026] Brian Connolly biography | https://sps.columbia.edu/person/brian-connolly
- [LinkedIn, 2026] Russell Scibetti profile | https://www.linkedin.com/in/rscibetti