In November, more than one in five failed.In May, they all looked the same.

A tray of kiwifruit leaves the packhouse and enters an inventory. From there someone has to decide where it goes and when: onto a chartered reefer sailing direct to Japan, into a container liner that calls at several ports on the way, or into coolstore to be held and sold later in the season. Each of those is a different amount of time between the packline and a person eating the fruit.

The tray does not care which one it gets. The fruit inside it does. Some lines will still be excellent after five months in store and a month at sea. Others are already carrying something that will express in a fortnight. On the day the assignment is made, both look identical, and the assignment is made anyway, because the ship sails on Thursday.

Follow one tray from the orchard. The fruit is picked, and the bins sit to cure, optimally for about forty-eight hours in a covered, ventilated space. That dries the stem scar and reduces the risk of botrytis later. Then it runs the packing chain, and this is where the industry does its sorting. In the industry’s own words, packing is the key control point where the fruit is segregated into market-acceptable product. It is graded for defects, sized, labelled, packed. Quality control staff sample it. It passes.

What the grade cannot catch is damage that has not surfaced. Zespri’s technical manager Frank Bollen put it plainly to growers: Most of the damage is almost impossible to see at the time of packing, but turns into high levels of rots, softs, non-pathogenic fungal growth, and superficial skin rub during storage and shipping. The same briefing recorded fruit that had been dropped onto other fruit or into bins taking five or more days to show damage, so couldn’t be seen during packing.

Five or more days. The grade takes one. Whatever happened during picking is already inside the fruit, and the packline is looking at it, and it looks fine.

Then the clock starts and never stops. Botrytis seeded at flowering waits for cold. Low-temperature breakdown surfaces after months. Bruising sits under the peel, optically masked. The tray goes into a coolstore where it may sit for weeks, then onto a vessel for two to four weeks, then into a distribution chain, then onto a shelf. At every one of those stages it is still passing, because there is nothing to see.

It fails in a supermarket in Shanghai in November, and comes back as a claim, and by then the fruit, the season and the decision are all gone.

The second option is worth reading closely, because it contains a forecasting problem stated as a commercial one. Sold for as long as it meets the quality expectations required is a question about how long New Zealand fruit will hold the shelf. Answer it well and New Zealand keeps the window. Answer it badly in either direction and you either ship fruit that becomes a claim, or you hand a month of the highest-value market in the world to a competitor who is already there. Chinese G3 now runs September to March, so Shanghai consumers effectively have twelve-month supply.

The length of the New Zealand selling window is now a forecasting output, not a marketing decision.

And the industry already has machinery that prices time. KiwiStart compensates growers for fruit sold early; Time Payments compensate for fruit sold late, and cover the fact that as kiwifruit is stored longer, it requires additional coolstorage and because the fruit is deteriorating overtime, condition checking, repacking, fruit loss, and taste compensation levels all increase. That machinery prices the cost of holding fruit. It does not tell anyone which fruit will survive being held.

The incumbent is not a rough heuristic. It is a governed, industry-endorsed system, and it works. Zespri’s Pool Policy Manual defines the mechanism precisely: the Sales Forecast is the allocation of available NZ sourced kiwifruit to Zespri market (country) by class, variety, growing method, based on a crop estimate. Market allocations of Class 1, Class 2, Non-Standard Supply and collaborative marketing volumes by market, product group and size are set out in it.

State its strengths, because they are real. It is auditable and pool-fair, so every grower can see the rules. It clears an entire national crop, on time, into dozens of markets, every year. It handles size and class, which are the attributes a customer contracts on. And it is backed by a system that reconciles right down to the tray: trays submitted, net submit, trays supplied, fruit loss reason codes, all defined.

Its limitation is not accuracy. It is that every variable in it is an attribute of the fruit on the day, and none of them is time. Class, variety, growing method, size and crop estimate are all known at packing and all static. Remaining shelf-life decides whether the allocation was right, and it is the one property the allocation does not carry.

The industry is not blind to time either. It manages time by hand, expensively, at the last possible moment. Zespri puts a Fruit Monitoring Technician aboard every chartered reefer from mid-March to mid-June, assessing fruit development, monitoring flesh and air temperatures, and releasing ethylene on request so fruit arrives Ready to Retail. That is a skilled person, on a ship, adjusting one voyage. It is a real intervention and it is the strongest evidence that the problem is understood. It is also as far downstream as an intervention can possibly be made: by then the lane is chosen, the fruit is loaded, and the only remaining variable is ripeness.

So the bar is this. Does a forecast of remaining shelf-life, made at packing, change which lane a line is assigned to? And does that change show up as fewer claims in the back half of the season, without forfeiting sales that would have been fine?