An experiment in reading public records: what does the data actually say about who owns a small upstate New York city? We pulled 161,000 parcels and 9,147 active LLCs to find out. Two strategies emerged — one visible, one invisible — both legal, both effective, both quietly running a meaningful share of the local economy.
A Question Hiding in Plain Sight
It started with a practical problem. Looking for a residential contractor in Jefferson County, New York — the kind of question every homeowner asks — turned out to be harder than it should have been. Local directories were thin. The chamber of commerce site lagged behind the actual market. Google Maps surfaced the same handful of corporate franchises that show up everywhere. We ended up paying corporate pricing for work that any number of local operators could have done for half the cost. If we could find them.
That experience prompted a different question. Not “where are the contractors,” but “what does the public data actually say about who is doing business in this county, and how?” We had reason to suspect that the visibility gap was structural rather than coincidental. So we pulled the data and looked.
This is what we found.
What the Data Actually Looks Like
Jefferson County, NY has 23 municipalities, a population of roughly 117,000, and a single dominant economic engine: Fort Drum, home of the 10th Mountain Division. The county contains the City of Watertown (about 25,000 people) plus 22 surrounding towns. Most of the rest of the population lives in those towns, including Le Ray, which hosts Fort Drum and is the largest single municipality by population.
New York State maintains a public parcel-data program. As of the 2025 publication, 38 counties have given the state permission to share their parcel records with the public. Jefferson County is not one of them. Its three neighbor counties — Lewis, Oswego, and St. Lawrence — all participate. Jefferson does not.
The City of Watertown publishes its own GIS data — 9,135 parcels with owner names, mailing addresses, and property classifications. That accounts for 21.1% of the county’s population. The rest of Jefferson County, including Le Ray, is structurally opaque to anyone trying to map the local economy from public sources.
This matters less than you might think, because the City of Watertown is the commercial center of the county. The downtown corridors, the highest-density apartment buildings, the bulk of the small commercial parcels — they’re all inside the city limits. Within that 9,135-parcel sample, the patterns are clear enough.
Layered on top of the parcel data, we pulled 9,147 active corporations and LLCs registered to addresses in Jefferson County from the New York State Department of State public dataset. That registry tells you who has filed a legal entity. The parcel data tells you who owns physical real estate. Crossing the two reveals what kind of business operators actually exist here, and what strategies they’re running.
Strategy A: The Vertical Stack
The first pattern that emerges from cross-referencing parcels with corporate filings is what we’ll call the vertical stack. It belongs to a single operator running a real-estate operation as a series of separately-incorporated LLCs, each isolating liability for one segment of the value chain.
One operator in our dataset has filed eight separate LLCs at the same Watertown address over a nine-year period. The filings, in order:
This is a textbook playbook from the real-estate investment world, executed precisely. Each LLC isolates one segment’s risk from the others. If a tenant lawsuit hits the rental holding company, the construction arm and the management company are not exposed. If a construction job goes sideways, the rentals are safe. The same individual signs every filing, but the entities themselves are legally distinct.
The pace is also notable. One new LLC every 12 to 18 months, expanding into adjacent value-chain segments rather than scaling within one. Started with management. Added rentals. Bought the building. Added a flipping arm. Then started doing his own construction. By 2024, this single operator was performing every step of the rental-property value chain in-house, billing each phase to one of his own companies.
From a public-data standpoint, the operator is fully visible. Eight LLCs with the same registered-agent address are easy to find once you know to look. But that requires looking at the corporate registry rather than at a local business directory — and most people, including most local residents, never do.
Strategy B: The Quiet Accumulator
The second pattern is the inverse. An individual who owns more parcels in the City of Watertown than the regional hospital does, holds every parcel in his personal name, and has filed exactly zero corporations or LLCs.
In our dataset, the largest non-government private landlord owns 23 parcels in the City of Watertown. By way of comparison: Samaritan Medical Center — the regional hospital — owns 20. This single individual is a more concentrated owner of city real estate than the largest healthcare institution in the region.
What does he own? Almost entirely residential rental properties:
Geographically, the parcels cluster. Six of his 23 are on a single street. Several others are within walking distance of each other. Three of the four vacant lots sit within 150 feet of his own home — effectively his back lot. This is not a portfolio assembled by an investment fund. It’s a portfolio assembled by a person who knew the local market, bought adjacent properties when they came available, and held.
The mailing address on every parcel is the same: his own home, which is itself one of the 23. He lives in one of his rentals. The other 22 he rents out.
From a public-data standpoint, this operator is almost completely invisible. There is no LLC name to search. There is no website. There is no business card. The only thing he has filed is the deeds themselves. A general Google search for his name returns no relevant results, despite the fact that he owns more residential real estate in the city than nearly anyone else not operating through a corporate entity.
What Both Strategies Reveal
These are two operators in the same city, at roughly the same scale, running fundamentally different strategies. One operates as an eight-entity vertical stack with a registered office. The other operates as a sole individual with no corporate shell at all. Both work. Both are legal. Both are quietly running a meaningful share of the local housing market.
The contrast points to several observations about how small-city economies actually function.
Visibility is a strategic choice, not a default. The operator running the vertical stack is highly visible to anyone who knows where to look — the corporate registry, the property records, the business address. The accumulator running properties under his own name is essentially invisible to anyone who isn’t already in the local network. Neither approach is hiding anything illegal. They’ve simply made different bets about whether being findable is a feature or a liability.
The “missing local economy” is mostly a directory problem. When residents complain that the chamber-of-commerce site is bad, or that there are “no local contractors,” what they mean is that the directories they consult don’t reflect the actual operator base. The data exists. Jefferson County has 9,147 active legal entities, 373 of them in construction or related trades. They aren’t missing. They’re unindexed.
Public records carry more signal than people realize. Crossing two free, publicly-available datasets — corporate filings and city parcel records — surfaces patterns that no one is publishing. The vertical stack only becomes obvious when you sort by registered-agent address. The accumulator only becomes obvious when you sort parcels by owner. Either question takes about ten lines of SQL to answer. Almost no one writes those ten lines.
A Public Directory of What We Found
The exercise produced something we kept. Crossing the corporate registry against the parcel data, classifying each entity by trade, and indexing it for search gave us a working directory of 9,147 active businesses in Jefferson County, broken into 22 trade categories — construction, plumbing, electrical, masonry, landscaping, restaurants, cleaning services, and so on.
We’ve published it as a free public preview at data.wbaconsulting.org. It’s browsable by category, searchable by name or address, and shows each entity’s registered office, filing date, and (where matchable) the Watertown parcels owned by that entity. The vertical-stack operator from this analysis appears on his own page, with all eight of his LLCs listed beside it. The accumulator does not appear, because he has no legal entity to list — which is itself a data point.
The directory is a work in progress. The largest gap: sole proprietors operating under a “doing business as” name file with the County Clerk rather than the State, and those filings are not yet included. Many of the one-person handymen, gardeners, and masons in the North Country operate this way, which means we currently undercount them. Closing that gap requires a records request to the Jefferson County Clerk, which is on our roadmap. We’ll publish what comes back.
What We’re Looking At Next
Three follow-ups are open. We’ll publish each as it lands.
The opacity question. Why has Jefferson County not opted in to the state’s public-parcel program, when its three neighbors have? We’ve drafted a polite request to the County Real Property Tax Services office asking for the rationale. Whatever the answer — if any — is itself a piece of the story about local government data transparency in upstate New York.
The foot-traffic layer. Parcel ownership tells you who holds the real estate, but not where the actual economic activity is. There are proxies for foot traffic that don’t require expensive commercial datasets: parcel turnover rates, building permit volume, business filing density by hex grid, OSM-tagged points of interest. We’re testing whether those can be combined into a “where to open a business” map for the North Country.
The sole-proprietor gap. The Jefferson County Clerk holds the records of every “doing business as” filing in the county. Adding those to the directory roughly doubles the operator population we can see. It would also let us answer questions like “how many handymen actually exist here” with something more than guesswork.
If you operate a business in Jefferson County and you’re miscategorized in our directory, or you’d like to be added, contact us. This is a free public service that exists because the data was sitting there and someone needed to read it.
For Jefferson County Specifically
This kind of analysis is exactly what WBA Consulting does. The directory is a free public artifact, but the methodology behind it — pulling and cross-referencing public datasets, building structured intelligence that didn’t exist before, productizing the result — is the work we do for clients.
If you’re a local business owner, a municipal office, a regional development organization, or a real-estate operator who wants to understand the actual shape of the North Country economy, the same approach applies to whatever question you’re trying to answer. We’re already deep in the data. The directory is the visible part. The custom analyses sit underneath.
Reach out at wbaconsulting.org/inquiries if there’s a question about this region you’ve been wanting answered and the public directories haven’t been enough. We’re based here. We read the data.
Related Analysis
- AI Agents Need Credentials Too: The Secret Management Gap Companies Aren’t Seeing
- The Next AI Upgrade Is the Environment, Not the Model
- AI Adoption vs AI Maturity: What Companies Get Wrong
Methodology note: Parcel data from maps.watertown-ny.gov (9,135 parcels, retrieved May 2026). Corporate filings from the New York State Department of State Active Corporations dataset on data.ny.gov (9,147 active entities registered in Jefferson County, retrieved May 2026). Municipal boundaries from NYS Civil Boundaries. All data is public; no scraping or paid sources involved. Trade classification is automated from registered-entity names and reaches roughly 39% coverage of all filings, biased toward operators with descriptive names. The remaining 61% are valid businesses with owner-name or generic LLC names that don’t disclose trade.