How to Start a Farm: The Complete 2026 Blueprint.

A farm in the United States does not have one customer. It has four distinct buyer classes with completely different economics, and choosing among them before you buy a tractor is the single most consequential decision you will make. The first class is the commodity buyer: the local grain elevator, the co-op, the ethanol plant, the packer. They pay board price minus basis, they take unlimited volume, and they do not care who you are. A commodity row-crop operation in this channel typically runs $200,000 to $2,000,000+ in annual revenue on 10% to 25% gross margin and a net margin somewhere between negative 5% and 12%. You are a price taker with zero marketing cost and zero pricing power.

The second class is the direct-to-consumer buyer: CSA subscribers, farmers market shoppers, farm stand traffic, on-farm pickup. Small direct-to-consumer vegetable farms typically land at $50,000 to $300,000 in annual revenue with 35% to 60% gross margin, but net margin still runs negative 10% to 15% because the labor and marketing costs that create the higher gross margin eat most of it. The third class is wholesale intermediate: restaurant chefs, food hubs, regional grocery produce buyers, and school or hospital food service. They pay above commodity and below retail, they buy on consistency and volume commitments, and they will drop you for one late delivery. The fourth is livestock and grazing, at $75,000 to $1,000,000+ revenue and 15% to 35% gross margin, where your buyer is a sale barn, a packer, or a freezer-beef customer buying halves and quarters.

Treat these revenue and margin figures as planning ranges, not published federal benchmarks. The federal source material available does not contain a single national farm profit and loss table, and any page that gives you one precise national number for farm margin is guessing. What is well documented is the cost side: the Farm Credit Administration names labor, livestock purchases, feed, seed, and fertilizer as the largest startup costs for new farmers and ranchers, and BLS puts crop farmworker pay at a $16.95 per hour median and $17.96 per hour mean. Build your plan from the cost side, which is knowable, and stress-test the revenue side, which is not.

Mixed crop and livestock small farms sit in the middle at $100,000 to $500,000 revenue, 20% to 45% gross margin, and negative 15% to 10% net margin. Notice that every one of these four models has a negative bottom of the net margin range. That is not pessimism, it is the structural reality of a business where you commit cash to seed, feed, fertilizer, and payroll months before a single dollar arrives, and where a drought, a price collapse, or a broken irrigation pump lands entirely on you.

Siting a farm is a four-variable problem and most first-time buyers optimize on the wrong one, which is price per acre. The four variables are soil, water, zoning, and distance to your buyer. Soil comes first for crops: pull the USDA Web Soil Survey for the parcel, look at the capability class, drainage class, and rooting depth, then confirm with your own composite tests at $300 to $5,000. A cheap parcel with a hardpan at 12 inches or seasonal high water table will cost more in tile drainage and failed stands than the land discount ever returned.

Water is the variable that kills deals late. Confirm well yield in gallons per minute against your peak irrigation demand, and confirm the legal right to use it. In the western states, water rights are separate property attached to the land and may be junior, seasonally curtailed, or already sold off. In much of the Midwest and East, you need a state withdrawal permit or groundwater district approval. Drilling and irrigation infrastructure runs $2,500 to $250,000+, and a well that comes in at half the yield you modeled means half the acreage you planned.

Zoning determines what you may legally do beyond growing. Growing is usually protected by right in agricultural districts and by state right-to-farm statutes. Selling is not. On-farm retail stands, commercial kitchens, u-pick, weddings, farm dinners, and seasonal worker housing frequently require a conditional use permit from the county planning office, which means a public hearing, neighbor notice, and weeks to months of delay. If your revenue model depends on customers driving to you, verify that use before you close, not after.

Distance to buyer is a hard cost, not a preference. Commodity farms measure it as basis: the difference between board price and what your local elevator pays, driven by trucking distance to the terminal. Direct-to-consumer farms measure it as market access, and a farm 90 minutes from a strong metro market pays for that in vehicle expense, driver hours, and product quality on arrival. Ground within 45 minutes of a metro area typically carries a land premium of a meaningful multiple over comparable rural acreage. That premium is often the correct purchase for a direct-sales model and the wrong one for a commodity model.

Farm unit economics only make sense at the enterprise level, meaning one crop or one livestock class at a time, with a cost per unit that includes everything from seed to the drive to market. Take a direct-to-consumer vegetable bed. Your variable cost stack is seed or transplants, amendments and fertility, drip tape and plastic, harvest labor, wash-pack labor, packaging, and delivery. On a 35% to 60% gross margin, a $4.00 retail bunch of carrots leaves roughly $1.40 to $2.40 after direct costs. The labor line inside that is the one that moves: at a $16.95 per hour crop farmworker median, a picker who bunches 60 units per hour puts about $0.28 of direct labor into each bunch, and one who bunches 30 puts in $0.57.

Now run the same product through wholesale. The same carrots move at roughly 40% to 55% of retail to a food hub or grocery buyer, but your packaging simplifies, your harvest labor per unit drops because you are pulling volume rather than filling a market display, and you eliminate the eight-hour market day where the owner earns nothing while standing behind a table. Many small farms discover their wholesale gross margin per hour of owner time exceeds their retail margin even though the retail gross margin percentage looks better on paper. Model both channels per hour of labor, not per unit sold.

Commodity economics work differently and more brutally. Your revenue per acre is yield times price minus basis, and you control only yield. With a 10% to 25% gross margin and a 2% to 5% swing in fertilizer or fuel cost, an ordinary input inflation year converts a profitable acre into a break-even acre. Livestock is slower still: a feeder animal purchased today ties up cash for 12 to 24 months, and the feed conversion cost, one of FCA's named top startup expenses, accrues daily whether the market cooperates or not.

The number nobody puts in the model is unsold product. A farm that grows 100 units and sells 78 has just added 28% to the true cost of every unit sold. Track sell-through by crop by week from your first season. Crops with sell-through under 80% are subsidizing themselves from your profitable lines, and the fix is planting less of them, not marketing harder.

At opening, most farms are one person: the owner-operator, who is simultaneously the production manager, the equipment mechanic, the bookkeeper, and the sales rep. That is workable on a very small leased operation and it is the reason the low end of the startup range sits near $40,000. It stops working the moment two production tasks compete for the same daylight, which usually happens in the first harvest week. Plan the second hire before you need them, because hiring during peak season means hiring whoever is available.

A small commercial farm typically runs the owner-operator plus one general farmworker plus seasonal labor at planting and harvest. BLS reports crop, nursery, and greenhouse farmworkers at a $16.95 per hour median and $17.96 per hour mean, or $35,260 median and $37,360 mean annually. Farm and ranch animal farmworkers run slightly higher at $17.17 per hour median and $18.02 per hour mean, or $35,710 median and $37,490 mean annually. Agricultural workers overall carry a $35,980 median annual wage, and the catch-all agricultural workers, all other category runs $19.53 per hour median and $40,620 annually. Budget above the median for anyone who can run equipment or supervise unsupervised.

A mid-size diversified farm adds a farm manager, a dedicated equipment operator, and packhouse or CSA support staff. A livestock operation needs an animal care worker with real stockmanship, equipment support, and seasonal help at calving, weaning, or shipping. A farm with direct sales or processing adds packing, delivery, and a compliance and administrative role that most owners try to absorb themselves and should not, because that is the role that keeps your pesticide applicator license current, your FSIS or FDA paperwork clean, and your crop insurance sales closing dates from slipping.

If you use H-2A, treat it as a separate business system, not a hiring channel. The Department of Labor's Adverse Effect Wage Rate framework governs pay, and the 2026 material shows a $15.96 per hour figure used in average AEWR computation for surety bonds. That is program-specific and not the same as ordinary domestic farm wages. The program also obligates you to provide housing, transportation, and a guarantee of three-quarters of the contract hours, and the certification process runs months. Start it long before your date of need or you will not have a crew.

You must sell before you harvest. The first 90 days is a sales window, not a growing window, and the farms that survive year one enter their first harvest with committed demand already booked. For a direct-to-consumer model, that means opening CSA share sales in winter or early spring with a clear pickup location, a share size, and a deposit that becomes your seed and fertilizer money. Twenty to fifty shares at $500 to $800 each is $10,000 to $40,000 of preseason cash that costs you nothing in interest. It also caps your risk, because you now know exactly how many units must exist.

For wholesale, the first 90 days is a list of named buyers and a physical sample. Build a list of 15 to 25 targets: independent grocers with a produce buyer, chef-driven restaurants, food hubs, school and hospital food service, and regional distributors. Contact them in the off-season, ask what they currently buy and from whom, and what they cannot reliably get. Then grow that. Chefs buy from farms that answer the phone and deliver on the promised day, not from farms with the best story. Bring product samples in the packaging you will actually use, and give them a price list with your delivery day and minimum order on it.

For commodity, the first 90 days is a relationship with your local elevator or co-op grain merchandiser plus a decision about forward contracting. Ask for the basis history for your delivery point, understand your storage options, and decide what percentage of expected production you are willing to price before it exists. Also register with your county Farm Service Agency office and establish farm and tract numbers, which gates access to FSA loan programs, disaster assistance, and crop insurance.

In all channels, get the physical presence right early. Farmers market vendor applications close months before the season and many established markets have waitlists, so apply in winter. Claim your listing on state department of agriculture producer directories and USDA's Local Food Directories, which cost nothing and drive real search traffic. Post a simple site with what you grow, where to buy it, and the pickup or delivery schedule. Marketing budget of $500 to $25,000 is a range that mostly reflects whether you are running a CSA with packaging and labels or selling into a commodity channel where nobody ever sees your name.