Welcome Homes

Snapshot

Welcome Homes is a New York-based direct-to-consumer homebuilder trying to turn single-family new construction into an online purchase. Buyers bring a lot — or pick one from Welcome's pre-vetted inventory — select from a short menu of architect-designed models, configure finishes in a browser-based design studio, and receive a guaranteed, all-in price that covers land preparation, the house itself, and a promised delivery roughly six months after permits are pulled. Founder and CEO Alec Hartman and his co-founders built the first iteration of the company in 2020 after a personal detour through self-general-contracting: Hartman couldn't find a new-construction home that fit his family in the New York suburbs, acted as his own GC to build one, and decided to productize what he'd learned. The DigitalOcean alumni team — Hartman, Mitch Wainer, and Ben Uretsky — raised a roughly $5.35M seed from Global Founders Capital in October 2020, launched in the tri-state area in 2021, and closed a $29M Series A led by Era Ventures that was announced in January 2023.

The pitch is deliberately Teslaesque: a car-configurator interface for houses, a fixed sticker price, no haggling with a general contractor, no six-month pricing merry-go-round between architect, bank, and township. By November 2023, Welcome had rolled out AI-powered land vetting — trained on thousands of in-house civil-engineering site assessments — that compresses what the company says is normally a three-week-to-three-month property due-diligence process into seconds. It paired that with a "Make an Offer" product that routes a buyer's bid to a network of local builders, which Welcome compares to a rideshare dispatch. The combined platform now lists more than 19,000 buildable, pre-vetted properties across more than 60 counties in the Northeast, Mid-Atlantic, and parts of the Southeast.

What it is and how it works

The product decomposes a new-construction purchase into three screens: pick a model, pick a lot, pick finishes. The model menu is intentionally narrow. Welcome launched with a wider custom range, discovered that choice paralyzed shoppers, and today sells principally two plans — the Oasis Flex (three bedrooms plus a flex room, roughly 2,818 square feet under air conditioning) and the Oasis 4 (four bedrooms at 2,670 square feet under air). Base pricing on the house starts at $550,000. The land layer sits on top: Welcome publishes pre-vetted parcels with their all-in price, from sub-$200,000 teardowns in Westchester to multimillion-dollar lots stretching past $15M. Buyers can also point Welcome at a parcel they already own and let the AI feasibility model confirm buildability.

Behind the configurator, Welcome doesn't own factories or crews. It is an orchestrator. Permits, land prep, septic, grading, excavation, and the home itself are quoted together as a single guaranteed price, financed by the buyer with a construction-to-perm loan that is interest-only during the build and converts to a traditional mortgage once the certificate of occupancy is issued. Welcome dispatches the work to a bench of local general contractors and specialty trades, and its engineering staff standardizes the plan and spec package. The company's claim is that algorithms absorb what a human project manager traditionally hand-holds — zoning checks, setback math, cost estimation — and that the GCs on the other side simply execute a well-defined scope.

Welcome has not disclosed homes closed or revenue publicly, and third-party databases list revenue as unreported. Geographies have ratcheted outward: launch in Westchester, Fairfield, and northern New Jersey in 2021; DC metro and Philadelphia in 2022; Maryland, Pennsylvania, Delaware, Virginia, Georgia, and North Carolina through 2023, with publicly stated ambitions to push further into Texas and the Sun Belt.

Why it might work

Three things are genuinely different about this model versus either a production builder or a traditional custom builder.

First, the customer problem is real and enormous. The U.S. single-family new-home market produced roughly 679,000 for-sale closings in 2025 (NAHB), with the top-ten builders controlling 43.6% of them. The remaining ~56% of new single-family closings — call it ~380,000 homes a year — are produced by a long tail of regional and local builders whose sales process is analog, opaque, and brutal for the buyer: spec homes you didn't design, cost-plus contracts that balloon mid-build, and timelines quoted as "six to nine months" that routinely slide into fourteen. A typical existing-home sale in the Northeast involves housing stock that is 60–90+ years old; in the DC metro the median single-family home is 75 years old, and in Philadelphia it is 93. That is Welcome's strongest narrative wedge.

Second, Welcome is attacking the part of the stack where production builders don't play. D.R. Horton and Lennar need large master-planned developments with hundreds of lots to pencil; they don't do infill, teardowns, or scattered-site custom lots. That leaves the lot-by-lot market structurally fragmented, which is exactly where software leverage compounds — one buildability model, trained once, can underwrite 19,000 scattered-site parcels where a Horton land-acquisition team would never bother.

Third, the fixed-price promise is the actual product. Anyone in homebuilding can tell you what a house cost after the fact. Almost no one can quote a buyer an all-in number before soil tests, without clawing it back later. If Welcome can hold that number — because the land is pre-vetted, the plan is standardized, and the trades bid against a well-defined spec — then the two-sided marketplace flywheel (more lots pull in more buyers, which pulls in more builder density, which tightens quotes) starts to look like a defensible business rather than a design magazine.

Why it might not

The construction-tech graveyard is instructive and crowded. Katerra burned roughly $2 billion of SoftBank money trying to vertically integrate design, factory, and GC, and shut in 2021. Veev, a former unicorn attempting prefab wall panels, failed to close financing and shut in late 2023. Cover Technologies, the Los Angeles backyard-ADU prefab darling, went dark around the same time. Mighty Buildings pivoted away from whole homes. ICON built a handful of 3D-printed houses with Lennar but remains a niche experiment. The pattern is clear: software works, but construction is a labor-arbitrage, site-specific, permit-bound business whose last mile cannot be venture-funded into submission.

Welcome has made a cleaner bet than Katerra or Veev — it doesn't own a factory or a workforce — but that cleanliness is also its ceiling. Because the trades are subcontracted, every project is only as fast and reliable as a local GC in Westchester or Raleigh on a given week. Reddit threads and real-estate forums have surfaced the same complaint that haunts the sector: timelines quoted at six months after permit that stretched to nine or twelve, and a limited model range that is take-it-or-leave-it. Fixed-price works until a builder walks or a soil report surprises, and Welcome has to eat the delta or renegotiate.

The configurator itself is also eminently copyable. Toll Brothers already lets custom buyers choose structural options online; Lennar's Next Gen tooling is competent; D.R. Horton's spec-home inventory is already discoverable on Zillow. A production builder that wanted to offer a lot-by-lot configurator for infill could probably ship one in a year. Welcome's moat isn't the UI — it is the land data, the pre-vetted parcel inventory, the pre-negotiated GC network, and the willingness to carry contract risk. Those take years to build and are easy to under-appreciate; they are also hard to point at in a pitch deck.

Finally, macro is unforgiving. Single-family starts fell 6.9% in 2025, permits fell 7.4%, and mortgage rates have remained a tax on new-construction demand. Welcome's product is most compelling in exactly the environment — rising rates and shrinking existing-home inventory — that also shrinks the pool of buyers who can finance a $900K build.

Funding and team

Welcome has disclosed roughly $34.4M across two rounds: a $5.35M seed in October 2020 led by Global Founders Capital (Rocket Internet's venture arm) with a wide syndicate of 19 investors including 8VC, Foundamental, Soma Capital, Arkin Holdings, Elefund, and angels Joe Lonsdale, Qasar Younis, and Josh Buckley; and a $29M Series A that closed in September 2022 and was announced in January 2023, led by Clelia Peters's Era Ventures with Parker89, Montage Ventures, Foundamental, Global Founders Capital, Activant Capital, Gaingels, Elefund, and Arkin Holdings participating. Peters — former president of Warburg Realty and a prominent proptech voice — joined the board. The founding team draws heavily from DigitalOcean: Hartman, Mitch Wainer (CMO), and Ben Uretsky, with Jesse Mauro as CTO. The company has publicly hired executives from Compass, Toll Brothers, Zillow, HGTV, and Google. Headquarters is 41 Madison Avenue in Manhattan.

Competitive set

Welcome overlaps uncomfortably with two camps. Among traditional homebuilders, the direct substitutes are the large publics — D.R. Horton, Lennar, PulteGroup, NVR, Meritage, Toll Brothers, Taylor Morrison, Dream Finders Homes, KB Home — plus the thousands of local custom builders who do one-off infill work. None of them sells online at a guaranteed price for a non-subdivision lot. Among construction-tech peers, the relevant names are Homebound (post-disaster and California custom builds), Juno (multifamily modular), Higharc (home-design software for builders), ICON (3D-printed homes), and Mighty Buildings — plus the cautionary tales of Veev, Cover, and Katerra. The clearest adjacency, arguably, is Homebound, which also combines proprietary design software with a GC network, but focuses on Californian rebuilds rather than a nationwide retail configurator.

Outlook: the open question

Can Welcome Homes industrialize custom home delivery faster than a production builder can copy the online configurator, or does the subcontractor model keep timelines too variable to sell a fixed price? The honest answer in late 2026 is that it is still a race. Welcome has the best shot of the current construction-tech cohort because it didn't take the factory bet that killed Katerra, Veev, and Cover; it is building the thinner, more software-shaped version of the same dream. But a fixed-price guarantee only works at scale if the builders on the other end of the dispatch hit their dates predictably, and that is not yet a solved problem anywhere in the United States. The $34M raised — modest by proptech standards — buys the company a few more geographies and a few more years to prove the physics. Watch two numbers: homes closed per quarter (not yet announced) and median permit-to-CO cycle time. If both trend the right way, Welcome has a real business and a pathway to Series B that is not a down round. If either stalls, the production builders will have all the time they need to copy the one thing buyers actually love about this company — the price tag that doesn't move.