On October 15, 2026, the elevator, the company that buys the grain, calls a Saskatchewan farmer. The example farm has 1,000 acres at 40 bushels an acre, so 40,000 bushels of canola sit in the bin and none are sold. During the call the farmer has to say how much to sell and at what price. The Western Producer reported that day that it isn’t unusual to see a 30 or 40 cent per bushel change in canola prices in a single day. On 40,000 bushels, one day’s timing is worth C$12,000 to C$16,000. The written selling plan, which grain extension guides call a marketing plan, settles that choice in advance on one page made before the first sale. It names the buyer and the portions (the parts of the crop sold one at a time). It gives the price or date that sends each portion to market, called its trigger, and the price you will judge the year against. It will not raise your price. Write it before you price the next bushel.
Does a written plan get a better price?
No, and the extension guide that supplies the worksheet below says so first. O’Brien wrote it for Kansas State University in 2000. He summarizes a study of Kansas Farm Management Association farms from 1989 to 1998, which asked which differences between farms repeated year after year. Its authors found that differences in no-till technology adoption, yields, cost of production, and profits were more persistent or consistent than differences in selling prices.
What kept one farm ahead of another was what it grew and what that cost. Differences in selling price lasted least. O’Brien concludes that a crop has to be efficiently produced before it can be effectively marketed. He adds that it is not an easy task for farmers to obtain better than average results in marketing their grain. The study is not evidence about planning, and he says so: the researchers assumed every farm marketed the same way and did not record whether any farm used a written plan.
The price still decides a lot in a year like 2026. The Western Producer’s Kevin Hursh, using Saskatchewan Ministry of Agriculture figures, reports that canola is the only Canadian Prairie crop expected to show a positive return this year. He adds that some producers will have sold canola for much more than C$17 a bushel at the farm, while others who priced early could be under that mark. Two farmers with the same crop can end the year on opposite sides of C$17, depending on when they priced.
One plan also cannot chase both a better price and a steadier income. O’Brien states the trade-off plainly: it is difficult if not impossible to consistently accomplish both of the goals of price enhancement and price risk reduction over time. Selling equal parts at fixed intervals gives up the chance of selling at the year’s low and the chance of beating the average. O’Brien does not defend it on price; such strategies may have merit if they are used to add structure and discipline to a marketing plan.
Write at the top of the page which of the two goals the plan serves. With no goal written, any result can be justified afterwards: sold early into a rally, it was managing risk; sold late into a slide, it was going for the price. The line costs nothing while none of the crop is priced. It is farm planning at its simplest, a rule written before the work starts so that anyone can check the work against it later.
What goes on the page?
Seven blocks, from the worksheet published with the Kansas State guide. You fill in the first five before anything is sold and the last two as the year goes on. The worksheet is a form, not advice, with a blank beside every line.
| Block on the Kansas State worksheet | What you write in it |
|---|---|
| Basic information | Name, crop, expected production, the period the plan covers, the date the plan was drawn up, five blanks for review dates, and what share of total sales is meant to go before harvest, at harvest and after it |
| Price goals | What the goal is based on (a better price, less price risk, covering cost of production or a profit target), and the goal itself as a price per unit |
| Market prices, outlook and price trend expectations | Futures and cash prices on a dated day, the place where each cash price was quoted, the factors expected to move prices, and in which direction |
| Marketing strategy | The pricing tool (a cash sale, a forward contract, futures or options), the quantity, the price goal and the time period for each sale |
| “Stop loss” price levels and contingency plans | The price level that forces the plan to change, and what replaces it when that level is reached |
| Evaluating performance | The goal, the price actually received, and the difference between the two |
| Comments on progress and performance | Dated notes, one per review |
The last two blocks come after selling. In the evaluation block you write the goal, the price actually received and the difference; the price received comes from adding up the settlement statements, not from memory. Doing that portion by portion is covered in the article on the sales against the plan.
Add one line the worksheet does not have: the buyer, at the top of the page, beside each product. With the buyer written down you can see how much of the year’s revenue depends on one buyer. Two products sold to the same buyer look like two sources of income and turn into one on the day that buyer stops paying. A second buyer is often the cheapest protection against that, since it changes nothing in the field, but it does little against price, because both buyers price off the same market.
In the third block, beside each cash price, the form leaves a blank for the place where it was quoted. It is the easiest line to skip. A cash price quoted by another elevator is not what you get at your own delivery point, so a trigger written without the place can go off at a price you cannot sell at. The expected production in the first block comes from a multi-year production plan, and the price the plan aims for has to match the price in the budget, or one of the two is wrong.
How do you split the crop into portions, and how much do you leave unpriced?
Three formats exist, and the page holds whichever you pick. Maples and a colleague at Mississippi State University Extension set them out for beginners in 2025. The target format sets four price targets, at the break-even price (the price at which sales just cover the cost of producing) and at 10, 20 and 30 percent above it. It prices 20 percent of expected production at each. The timed format prices the same 20 percent portions on four fixed dates, the first week of each of the four months before harvest. Those months follow Mississippi’s crop calendar, so a farmer elsewhere counts back four months from the farm’s own harvest. The hybrid uses the targets and falls back to the date for any portion whose target was not reached. In all three, 80 percent of expected production has a price by harvest.
All three leave at least 20 percent unpriced, to account for production uncertainty due to weather or other risks, helping to avoid the financial consequences of being over-hedged in a lower-yielding year. Over-hedged means priced for more bushels than the field delivers. Pricing everything quietly assumes a yield nobody has measured yet.
Each trigger is a goal with a number and a deadline, applied to one sale. A trigger written as “when the price is right” fails the way a goal written as “grow more” fails. Someone has to read it at nine in the morning and know whether it went off.
The example farm uses the hybrid format for next year’s crop of 40,000 bushels. The prices and dates are example values, not recommendations.
| Portion | Bushels | Trigger | Fallback date | Who may act |
|---|---|---|---|---|
| 1 | 8,000 | C$17.00 a bushel at the farm | First week of May 2027 | Owner |
| 2 | 8,000 | C$17.50 a bushel at the farm | First week of June 2027 | Owner |
| 3 | 8,000 | C$18.00 a bushel at the farm | First week of July 2027 | Manager |
| 4 | 8,000 | C$18.50 a bushel at the farm | First week of August 2027 | Manager |
| Unpriced | 8,000 | None; the owner decides once the yield is known | None | Owner |
Which price do you judge the year against?
The benchmark, the price you compare the year’s sales with, has to be chosen before the first sale. Chosen afterwards, it tends to flatter whoever chose it. Dietz and three colleagues, in the Journal of Agricultural and Applied Economics in 2009, studied wheat sold by Illinois and Kansas farmers from 1982 to 2004. The same sales scored three ways depending on the price they were compared with. Against a 24- or 20-month benchmark they came out about equal to the market. Against a 12-month benchmark they were slightly above the market. Against the harvest benchmark they were significantly less than the market. Nothing about the selling changed between the three results; only the benchmark did.
The study also tested the oldest line in the grain trade, that farmers sell two-thirds of the crop in the bottom third of the price range. The prices farmers received landed in the middle third of the crop year’s range, not the bottom third, between half and three-quarters of the time. The problem is less that farms sell at the bottom than that they are judged afterwards against whichever benchmark is at hand.
The good price a farmer waits for also moves. Mattos and a co-author ran a selling exercise on paper in 2012 with 75 grain farmers in southern Manitoba and asked at every step what price the farmer was holding out for. The answer was shaped mostly by the average of the last two prices the farmers had seen and by the highest price reached so far in the exercise. The target price drifted down a little each month. The 75 were volunteers, the sales were simulated on made-up price series, and real prices were climbing in those weeks. The result describes how the target price behaves, not how much anyone lost.
Write the benchmark in the second block of the worksheet, on the line for what the price goal is based on. It is an easy line to leave empty.
Who acts on a trigger, and when is the plan checked?
One named person acts, and the Kansas State worksheet has no line for the name. It has a place for the price level that forces the plan to change and what replaces it, which settles what gets done but not who does it. Write the name beside each trigger: the person who can act without calling anybody first. A trigger that needs a meeting is only a suggestion, and the market has moved by the time the meeting happens. Without the name, the decision goes back to the phone call. A written plan is one tool of farm management, and it works only when someone has the authority to act on it.
The worksheet has five blanks for review dates, after the date the plan was written. O’Brien explains that naming dates is meant to encourage sellers to periodically review and update their marketing plans to allow such plans to become dynamic decision-making processes rather than onetime exercises. A plan left alone may provide poor direction to marketers or become irrelevant in light of changing market conditions. A review date slips when nobody is named to call the review, so write that name too.
Where to start
An hour and a half, with the sales records of the last two seasons on the table. None of it needs a broker, a platform or a price forecast.