Let me start with a bill. Last fall, our warehouse wrote off a batch of yogurt: more than 180,000 yuan in goods, plus another 20,000 or so in disposal fees. The yogurt was not unsellable. Our batch and expiry management had simply failed to keep up — the system recorded the receiving date but not the production date or the shelf life, so pickers grabbed whatever was closest and the oldest stock ended up buried at the back. This is a course no food warehouse can skip. You either pay the tuition in daily discipline, or you pay it on the write-off slip.
A food warehouse does not hold piles of boxes. It holds hundreds of batches, each with a birth date and an expiry date. Which batch came in first, which one expires soonest, which one must be locked down if something goes wrong — if you cannot answer those three questions, your cycle counts can be perfect and your books can still be a mess.
FIFO Is Not Enough for Batch and Expiry Management
First-in, first-out sounds right, and in a food warehouse it regularly fails. Take one product in three batches with shelf lives of 12 months, 9 months, and 6 months. The batch that arrived first has the longest life; the one that arrived last expires soonest. Pick by receiving order and the short-life batch sits until it expires. The correct rule is first-expired, first-out: whichever batch expires first ships first, regardless of when it arrived.
We learned this the expensive way. In the summer of 2024, a seasoning client had two batches that arrived a month apart but carried shelf lives three months apart — the supplier had changed the formula and cut the shelf life from 18 months to 12, and nobody told the warehouse. The system picked first-in, first-out, and the short-life batch sat in a corner for four months. By the time anyone noticed, it had 40 days left, and the client had to clear it at a discount, losing more than 60,000 yuan. After that we made a rule: receiving must record the production date and shelf life, and any goods missing those two fields do not go onto the shelf. A shelf-life change needs written confirmation; a verbal heads-up does not count.
Here is a counterintuitive point: the longest shelf life is not necessarily the safest stock. Goods with a 24-month life turn slowly and can still expire; goods with a 3-month life turn fast and rarely cause trouble. Managing expiry is really about managing turnover speed, not shelf-life length. Do the math per SKU: divide monthly sales by inventory to get days of cover, then subtract transit and putaway time from the shelf life. What is left is the number of days you can actually sell. Run that calculation and plenty of stock that "looks fine" turns out to be standing at the edge of a cliff.
Batch records earn their keep a second way: when something goes wrong, they let you lock down precisely the affected goods. In 2023, a biscuit brand found a problem with one batch of raw material, and the client called us at midnight demanding to know, within two hours, which stores had received that batch and how much was still in our warehouse. Because receiving had recorded batch numbers and outbound destinations, we pulled the list in 40 minutes: 1,200 cases in the affected batch, 800 already shipped to 17 stores, 400 still in stock and locked on the spot. Without batch records, the only option would have been to stop all shipments and search case by case, and the loss would have been at least three times larger. Batch management looks like extra paperwork on an ordinary day. On the day something goes wrong, it is the thing that saves you.
Near-Expiry Disposal Is the Last Bill in Batch and Expiry Management
Near-expiry goods suffer from two kinds of mishandling. One is hiding them and reporting nothing until they expire. The other is panic-discounting everything and blowing up the price structure. Our warehouse now works to a fixed schedule, written into policy and followed by the book:
| Days of life left | Action | Who decides |
|---|---|---|
| More than 90 | Sell normally, pick by expiry first | Warehouse, by expiry picking |
| 60–90 | Alert the client, discuss promotion or transfer | Client confirms |
| 30–60 | Promotional clearance, discount confirmed in writing | Client confirms in writing |
| Under 30 | Stop routine shipping, separate approval required | Client plus warehouse manager, dual sign-off |
The table looks rigid, and that is the point: responsibility is clear. The near-expiry alert goes out 90 days ahead, not 30 days when it is already a rescue mission. Who receives the alert, how many days they have to reply, and what happens if nobody replies are all written down. The worst outcome is an alert that nobody answers while the goods age on the shelf and turn into a write-off slip. Almost as bad is mixing near-expiry stock with normal stock in the same location, so a picker grabs the wrong case — the near-expiry goods stay put while the fresh ones ship, the books still balance, and the expiry profile is already a mess. Once goods hit 60 days, we move them to a separate zone. Physical separation beats a system reminder.
Disposal itself is a bill. At the end of 2023 we disposed of a batch of imported biscuits worth more than 90,000 yuan: photographs on file, two-person count, a destruction certificate archived, and over 10,000 yuan in handling fees. The client asked whether the goods could be donated instead, and my answer was to follow the applicable rules — for the compliance position on food disposal and donation, refer to the latest official policy rather than deciding from experience. The destruction certificate matters because it is the only proof that the batch was genuinely destroyed and did not flow back into the market.
Run the Numbers Monthly, or the Numbers Run You
How well batch and expiry management is working does not need guesswork. Run three numbers every month: near-expiry value as a share of total inventory — the value of stock expiring within 90 days divided by total inventory value; the write-off rate — the value written off in the month divided by the value shipped that month; and batch accuracy — the share of picks where the batch picked matches the system record, checked by sample. If near-expiry value passes a tenth of inventory, purchasing and sales are already out of sync. If the write-off rate climbs two months in a row, stop and find the cause: overbuying, slow sales, or wrong expiry data at receiving.
There is a hidden bill most people never calculate: near-expiry goods occupy locations that good sellers could use. One pallet position holding stock that will be written off in 30 days is one pallet position less for stock that turns normally. When space is tight, that opportunity cost is larger than the write-off itself. So the near-expiry alert should go to the warehouse as well as the client — free up the locations early, schedule the promotion early, and do not wait until write-off day to discover the locations are jammed too.
These three numbers have one more use: leverage in a negotiation. When a client complains that storage fees are high, show them the write-off bill — how much write-off you saved them last year, and what that saving was worth. A storage fee looks very different next to a number like that. Avoided write-offs are earned profit. In a food warehouse, that is literally true.
In the end, batch and expiry management in a food warehouse is not about how expensive your system is. It is about whether the production date and shelf life were recorded correctly in the one minute at receiving, whether somebody answers the alert, and whether near-expiry goods have a written disposal schedule. Get those three things right and the write-off slips get thinner. Get them wrong and the most expensive system in the world just keeps a tidier record of the same mess.
By Lao Zhou, Logistics Cost Analyst







