A number with a date is a goal. A wish to grow more is an attitude, and the two behave differently under measurement. Locke and Latham, summarizing 35 years of research in the American Psychologist in 2002, reported that specific, difficult goals consistently led to higher performance than urging people to do their best, and the gap turned up in every set of studies they pooled: it ran between 0.42 and 0.80 on the scale that literature uses to size a difference, where 0.2 counts as small and 0.8 as large. Their sentence for why is short and unkind: when people are asked to do their best, they do not do so.
The reason is mechanical rather than moral. Do-your-best goals have no external referent and thus are defined idiosyncratically, which means every person in the operation quietly supplies their own version and every one of them is then right. Ohio State University Extension puts the same point from the management side: goals are the standard by which we will measure our progress, and a standard that each person defines privately is not one. That is the sense in which the planning axis uses the word: written before the work starts, and the same for everybody who reads it.
How wide the evidence actually is
Wider than any farm-specific study could be. Locke and Latham report that specific difficult goals have raised performance across well over 100 different tasks involving more than 40,000 participants in at least eight countries, in laboratory, simulation and field settings, with time spans from one minute to 25 years, and with effects found for individuals, groups, organizational units and whole organizations.
That breadth is the reason the finding transfers to an operation nobody in that literature ever visited. It is also the reason to be careful about what transfers. Goal difficulty raises performance until the limits of ability were reached or commitment to a highly difficult goal lapsed, which is a ceiling, not a license to set any number high enough to sound ambitious.
What specificity buys, exactly
Less spread, not just more effort. Locke and Latham are precise about it: where performance is fully controllable, goal specificity does reduce variation in performance by reducing the ambiguity about what is to be attained. The number does not only pull the average up. It pulls the range in.
Wherever the same job gets done more than once, by two people on one day or by one person on twenty days, the range is usually the expensive part. Two operators averaging the target with one well above it and one well below it costs more than two operators sitting on it, and an average is exactly what a vague goal produces, because each person filled the ambiguity differently and none of them was disobeying anything.
The five letters, and the two nobody agrees on
SMART is not one acronym. Two extension services publishing guidance for farms expand three of the letters the same way and two of them differently, which is worth knowing before a farm treats the checklist as settled.
| Letter | Ohio State University Extension, ANR-45 | Rutgers NJAES, FS1263 |
|---|---|---|
| S | Specific, focus on a specific problem or need | Specific |
| M | Measurable, some means of tracking achievement | Measurable |
| A | Action-oriented | Achievable |
| R | Realistic, aim high but stay possible | Relevant |
| T | Timed, goals are only useful when current | Time-bound |
The A and the R are doing different jobs in the two versions. Action-oriented asks whether anyone can start on Monday. Achievable asks whether the number is reachable at all. Relevant asks whether hitting it would matter. A farm that wants all three has to ask all three. Pick one published list, stop there, and the goal that comes out can be perfectly reachable while mattering to nobody and having no Monday to start on.
The letter most farms keep in name only
The M, because measurable and reportable are not the same requirement. Locke and Latham are unambiguous that for goals to be effective, people need summary feedback that reveals progress in relation to their goals, and that without it they cannot adjust effort or strategy. Their example is a goal to cut down 30 trees in a day, where people have no way to tell if they are on target unless they know how many trees have been cut.
A goal that can only be settled when the accounts close is measurable and gives nobody anything to steer by for eleven months. The gap is not hypothetical. In the Purdue University survey of 403 American producers in April 2023, 50.6 percent of farms used financial ratios to make decisions and 49.6 percent documented standard operating procedures for repetitive and routine tasks. For roughly half of those operations the channel that would carry the summary feedback does not exist yet, and setting a sharper number changes nothing until it does.
Locke and Latham also list what makes goal-setting fail to replicate at all, and the list reads like a description of a farm meeting: not providing feedback, not getting goal commitment, not measuring the person’s personal goals, not conveying task knowledge. None of those is a fault in the goal. All of them are faults in what happens after it is written, and the thing that decides whether any of them gets caught in time is a scheduled review date. Whether they are present on a given farm is settled in one afternoon by asking each person separately what the goal is, which returns a count rather than an impression.
The deadline is not decoration
It changes the pace of the work, on its own. Locke and Latham report that tight deadlines lead to a more rapid work pace than loose deadlines, in laboratory and in field settings alike. Ohio State Extension adds the expiry side of the same idea, that goals are only useful when they are current.
A research bulletin from Embrapa Cerrados, the Brazilian federal agricultural research corporation, ties the horizon to what the organization actually holds. Its three brackets are short term with resources already in hand, medium term with resources from the next budget, and long term with resources that lie beyond the coming year. The horizon stops being a preference and becomes a question about where the money comes from, which is the question that settles whether the goal belongs to this cycle or the next one.
Rutgers NJAES shows what the finished sentence looks like when both the figure and the date are actually in it: By December, Sunshine Farm will grow 2 acres of gourmet garlic and sell 12,000 lbs. of product through direct-retail at farmers markets, earning an annual gross income of $48,000 and an annual net income of $18,000. The figures are US dollars, the 2 acres are about 0.8 hectares and the 12,000 pounds about 5,440 kilograms, and those three readings are ours and not the fact sheet’s.
The example is American end to end, and what is worth copying from it is the count of fields and not the figures: a reader anywhere else swaps the unit, the currency and the month for the ones their own invoices and contracts already carry. No local figure appears on this page, because Rurivia does not invent a yield or a price. Area, volume, channel, gross and net, and a month. Any one of those missing and somebody on the farm gets to decide it later, alone.
Who owns the number between now and the date
One named person, with a review booked before anyone leaves the room. Stephanie Plaster, writing for University of Wisconsin-Madison Extension, sets out the conversion from strategy to action in eight steps, running from picking the top three strategies through to adjusting goals and steps as needed. Two of the eight are harder to hold than the rest, and that ranking is ours rather than a failure rate anybody measured: identify a person responsible for each step and determine a realistic timeframe for the goal to be completed, with the scheduled progress review immediately after them.
Adjusting as needed is the part that closes the loop rather than reopening the argument. A goal missed by 8 percent with a recorded reason and a named correction is a working control, and the step on which the four functions of management actually end. The same goal missed with no reading taken until the year ends produces a discussion about whether it was ever realistic, which nobody can settle, and which is where the next set of goals gets written vague on purpose.
A systematic review published in Heliyon in 2023, drawing on 24 studies published between 2013 and 2022, found that clear business goals and plans are linked to better farm results, and in the same table declines to call the link between recordkeeping-based planning and performance conclusive. Association, measured on farms. The mechanism underneath it was measured somewhere else entirely, on 40,000 people who were not farming.
Where to start
Take the vaguest thing the farm says it wants and force five fields onto it: quantity, unit, calendar date, the record the reading comes out of, and the name of whoever reports it.
A goal stated in money needs one field more than those five. The invoice supplies the volume, and the currency figure exists only because of a declared price assumption somebody made and almost never wrote down. A goal stated as profit left over needs the field after that one as well, because the subtraction that produces it has to be the same subtraction both times, which is what a declared margin target writes down before the cycle starts.
The goals themselves come out of somewhere, and it should be the written SWOT analysis rather than last year’s number with a percentage added to it, with written mission, vision and values deciding which of the candidates is worth the cycle. Here is the claim to argue with, and a farm with a year of readings behind it is entitled to argue: most farms do not have a goal-setting problem at all. They have a reporting-interval problem wearing a goal-setting costume.
Until the reading arrives more than once a year, sharpening the number is work done on the wrong end of the thing, and the sharper number will be discovered to have been missed at exactly the same moment as the vague one.