Keeping the harvest team on the same numbers

Harvest teams rarely fall out over the work. They fall out over information — or rather, over the gap between what one person knows and what everyone else assumes. The combine keeps cutting because nobody told it shed 2 filled at half three. The cart driver writes down the field name he thinks he was in. The office discovers in November that forty tonnes of Skyfall went on the feed heap in August, and nobody can say which day.

The fix isn't more talking. It's making sure the few numbers that matter live in one place everyone can see.

The three failures that keep repeating

They're the same on most farms, most seasons:

The combine doesn't know the shed is full. The store fills mid-afternoon, the person at the pad improvises a Plan B, and the next three trailers get tipped somewhere that made sense at the time. The records catch up later, or don't. The decision that should have been made at 300 tonnes — where does this crop go next — gets made at 400, in a hurry.

The cart driver guesses the field. He picked up from the combine somewhere in the middle of a 40-hectare block that used to be three fields. He writes "Barn Field" because it's probably Barn Field. Multiply by a season and the field yields — the numbers next year's cropping decisions lean on — are confidently wrong.

The variety mix-up surfaces in November. Two wheats, two sheds, one catchy Saturday with a relief driver. The blend is invisible until a merchant's sample disagrees with the declaration. With milling premiums commonly £10-£30/t over feed, 150 tonnes blended away is a £1,500-£4,500 mistake — and by November, unfixable.

None of these are caused by laziness. They're caused by information sitting in one head while a decision gets made in another.

Who needs to know what

Four roles, whatever the actual headcount — on plenty of farms one person covers two of them:

Role	Needs to know	Should be telling everyone else
Combine	Which field next, moisture readings, when to stop	Tank fulls cut, roughly where in the field, anything odd (lodged patches, green corners)
Cart	Which field they're serving, which shed each load goes to	Each load: field, crop, destination — recorded, not just remembered
Pad / drier	What's arriving, moisture per load, space left per shed	Shed levels, drier backlog, "this store is full at roughly X more loads"
Office	Contract specs, movement bookings, running totals	What's booked to move tomorrow and which shed it loads from

Read the table by columns and the problem is obvious: nearly everything one role needs to know is something another role learned hours earlier. Harvest organisation is mostly the plumbing between those columns.

Why radios and WhatsApp half-work

Radios and a WhatsApp group are better than nothing, and every farm uses some mix of both. But they share a flaw: they broadcast moments, not state. A radio call tells whoever was listening that shed 2 is nearly full. It doesn't tell the relief driver who starts after tea, and it doesn't exist at all by the time the office is reconciling loads on Sunday.

WhatsApp is a bit better — at least there's a scrollback — but a group chat is a haystack. "How much is left in the barn?" is answerable only by scrolling past forty messages about bacon rolls, and the answer you find is whichever guess was typed most recently, not the truth.

Both tools are fine for coordination — "beans next, not the wheat" — and poor as a system of record. The failure mode is using them as one.

A single live record beats broadcasting

The alternative is boring and effective: every load gets recorded once, at the time it happens, against its field, crop and destination shed — and everyone reads the same running totals instead of asking each other. The mechanics are covered in how to track harvest without a spreadsheet, but the organisational effect is the point here:

• The pad doesn't announce shed levels; anyone who cares can look.
• The cart driver picks the field from a list instead of free-styling a name.
• The office sees the day build in real time instead of decoding a group chat.
• The relief driver starting at 6pm inherits the actual state of play, not a verbal handover.

Questions stop being interruptions. That alone is worth having in the second week of wheat, when everyone's temper is two hours of sleep short.

Getting there is mostly a pre-season job — agreed field and shed names, tares taken, everyone shown the system — which is exactly the pre-harvest checklist territory.

The five-minute evening debrief

However good the live record, hold a short debrief while the day is fresh — five minutes over the last cup of tea, not a meeting:

• Do the day's loads look right? Any load someone isn't sure about gets fixed now, while memory is real.
• Where does the combine start tomorrow, and does the cart know?
• What's moving tomorrow — lorries booked, which shed, who's loading?
• Any kit problem that becomes tomorrow's bottleneck? A slow trailer hydraulic at 9pm is a note; at 11am it's a stopped combine.

Pair it with a glance at the five numbers worth checking every evening and tomorrow mostly runs itself.

Frequently Asked Questions

How do you organise a harvest team?

Define the four roles — combine, cart, pad/drier, office — agree who covers each, and make sure the information each role generates gets recorded somewhere the others can see it. Most harvest chaos is one role not knowing something another role learned hours earlier.

What's the best way for a harvest team to communicate?

Radios or a group chat for in-the-moment coordination, plus a single shared record for anything that needs to persist — loads, fields, shed levels. Broadcast tools half-work: they inform whoever was listening at the time and nobody else, which is why relief drivers and the office end up guessing.

Why do variety mix-ups happen at harvest?

Usually a shed decision made under pressure that never reached everyone — a store fills, someone improvises, and a trailer of one variety tips on another. They surface months later when a merchant's sample disagrees with the declaration, and by then the blend can't be undone.

Is an end-of-day harvest debrief worth it?

Yes, kept to five minutes. Confirm the day's loads look right while people can still remember them, agree tomorrow's starting field and movements, and flag any kit niggle before it becomes a mid-morning breakdown. It's the cheapest fix in the whole operation.