An agricultural business runs on a mix that few other sectors know: physical production tied to season and weather, administration around manure, crop protection and subsidies that leaves little room for error, cultivation or livestock decisions that rely on experience and observation, and a sales side that often runs through a limited number of buyers or cooperatives. In addition, the owner is usually out on the yard, in the barn or in the shed personally, and a large part of the knowledge about land, animals and machinery is not written down anywhere but exists in the head of one person.
That combination drives the outcome when a buyer comes to look. A buyer does not count the hours currently going in, but calculates what will keep running if the current owner falls away. In a business where administration, planning and even part of the cultivation decisions run on systems and recorded rules, more remains than in a business where everything sits with one person's feel for the work.
The shift runs through the business in parts, not evenly across everything. Three types of work can be distinguished.
Work AI can take over: filling in manure accounting and subsidy applications based on fixed rules, compiling reports for the bank or accountant, flagging deviations in sensor data from barn or greenhouse, and planning cultivation schedules based on historical yields and weather data.
Work that partly shifts, with human oversight approving or rejecting with reason: crop protection advice where a system makes a proposal but the grower decides based on the plot and the current state of the crop, or feeding schedules that a model calculates but that the livestock farmer adjusts based on what he sees in the animals.
Work that remains human work: assessing an animal's health through smell, behaviour and touch, negotiating with a buyer over price and delivery, and judging a harvest moment where land, crop and market weigh together in a way that cannot be captured in a few variables.
The difference between businesses lies not in the sector but in how the business is set up. A business that already works with recorded protocols, sensor data and digital administration can hand that first category over to a system now. A business where everything sits on paper and in the owner's head still has that step to take before AI can take over anything there.
As work shifts to systems, the weight of every driver a buyer takes into account shifts along with it.
Owner dependency weighs most heavily. A business where the owner is the only one who knows which plot needs fertilising when, or which buyer accepts which quality, represents a risk for a buyer that comes back in the price. If that knowledge is recorded and partly automated, that picture changes.
Profit quality also changes character. A saving that arises because one subsidy scheme happens to work out favourably, or one supplier offers a low price, counts differently for a buyer than a saving that sits in the process itself, for example because planning or administrative work structurally takes less time. The first disappears as soon as the circumstance falls away, the second continues to exist under a new owner.
Client concentration, the quality of the record-keeping in management information, and the predictability of yield also weigh in, and shift as more work demonstrably keeps running independently of the owner.
Regarding what this means for staff, the following applies: which steps an employer takes in response to changing work is up to the employer, along with the employer's own statutory requirements. This is not staffing advice and not grounds for a dismissal decision.
The agricultural sector is not the only one where AI shifts the weight of value drivers, and the comparison makes visible what is sector-specific and what applies everywhere. In the recreation industry, seasonal dependency plays a role comparable to weather and the growing season here, discussed via what makes seasonal businesses valuable to a buyer. In financial services, the emphasis lies instead on regulation and file-building, worked out in what regulation-driven work means for the sellability of an advisory firm. And in the cleaning industry, where labour forms a similarly large share of costs as in livestock farming, it becomes visible how labour-intensive service businesses build or lose value.
The underlying question is the same in every business: which work here can genuinely be taken over by AI, and which work cannot. That question is answered task by task with the work scan from FTE TO AI, applied to your own business rather than to the sector in general.
Anyone who first wants to know how their own management information scores on completeness and currency will find that worked out in how management information professionalises from loose Excel files to a system a buyer trusts. And anyone wondering what specifically depresses the price of their own business can read about that in the seven most common reasons a business turns out to be worth less than the owner thinks.
The eight drivers a buyer weighs are not all equally visible from the yard. The free value check consists of eight short questions, one per driver, and gives a first picture of which driver is depressing your business's price the most today. The full value scan, with a maturity score per driver, evidence base, owner dependency index and a two-year calendar towards the exit moment, is under construction.