What is the difference between shelf life and PAO?

Shelf life is the unopened stability period of a product — how long it remains within specification in a sealed pack. PAO, or period after opening, is how long it remains usable once the seal is broken, and it is shown as the open-jar symbol with a number of months. They measure different things and are not interchangeable. In the EU the distinction also determines what appears on pack: products with a shelf life under thirty months carry a minimum durability date, while those above it carry a PAO instead. Since a planning ceiling is computed from remaining shelf life, using the PAO figure produces the wrong number.

How does a retailer's date-code policy limit weeks of supply?

Most retailers will not accept stock without a stated share of its shelf life still remaining at delivery. That reserved share is unavailable to you, and so is the time the product spends in transit and in your own warehouse before it ships. What is left is the maximum useful cover you can hold — the true ceiling on weeks of supply for that item, independent of what demand or service targets would otherwise suggest.

Does a dating ceiling change how much safety stock you can carry?

It caps it outright. Safety stock is cover held against demand and supply variability, and the usual method sizes it from service level and lead-time variance without any upper bound. A dating ceiling imposes one: cover you cannot legally deliver is not cover. Where the calculated safety stock exceeds the ceiling, the shortfall has to be solved with responsiveness — shorter production lead time or more frequent smaller batches — rather than with inventory.

Why does a wide shade range end up with expired stock?

Because even batch sizing across uneven demand does it mechanically. If every shade is produced in the same run size, the slow shades hold far more weeks of cover than the fast ones at the same unit count, so they reach the dating ceiling first. The wider the range, the longer the slow tail, and the more reliably it expires. This is not a forecasting failure — it is what even batches do to uneven demand.

When should stock approaching its date be exited?

On a date computed backwards from the clearance channel's own intake requirement, not on a point in the sell-through curve. Below a dating threshold the stock cannot be sold into the main channel at any discount, so the usual markdown logic — lower the price until it moves — stops applying. Off-price and secondary channels have their own remaining-life requirements and their own lead times, so the exit has to begin early enough to clear those. Waiting for sell-through to signal a problem means arriving after both routes have closed.

When Dating Rules Cap Your Weeks of Supply

Q: How do you avoid expiring the slow tail of a shade range?

Not with smaller batches alone, because the filler has a minimum viable run below which it will not produce. The mechanism the industry uses is postponement: hold shared bulk or an uncoloured base and blend or fill to shade late, close to demand. That moves the commitment point from the finished shade to the base, so the minimum run applies to something every shade consumes rather than to each shade separately.

Shelf life, PAO and retailer date-code policy put a hard ceiling on how much cover you can hold. This guide covers computing that ceiling, why it caps safety stock outright, how a wide shade range produced in even batches converts mechanically into expired tail stock, and why the exit trigger is a date rather than a sell-through point.

Where this sits, and where it stops

A quick scope note, because it matters. Lot codes, batch traceability and expiry tracking belong in ERP and warehouse systems, and the health and beauty planning guide says so explicitly. Nothing here changes that.

The planning system's job is narrower: hold the dating ceiling as a constraint, and flag the items approaching it. It does not track lots. It makes sure the plan never asks for more cover than the product can legally deliver — which is a planning question, and one that standard cover logic gets wrong by default because it has no concept of an upper bound.

Computing the ceiling

Standard weeks of supply answers "how much cover do I want?" Dated product forces a prior question: how much cover am I allowed?

The arithmetic is subtraction:

Maximum useful weeks of supply = total shelf life − the retailer's reserved remaining-life share − time in transit and in your own warehouse before it ships.

Each term is a real deduction:

The reserved share. Most retailers require a stated proportion of shelf life still remaining at delivery. Stock below it is refused at intake. That share is simply unavailable to you.
Transit and warehouse time. The clock starts at production, not at your warehouse door. Everything between manufacture and the retailer's receiving bay is consumed from the same budget.
What remains is the ceiling — the most cover the item can carry regardless of what demand or service targets suggest.

Get the first term wrong and everything downstream is wrong, which is why the shelf-life-versus-PAO distinction is not pedantry. Shelf life is unopened stability; PAO is the period after opening. They answer different questions, and only the first belongs in this calculation.

The ceiling caps safety stock

This is the consequence most likely to be missed, because safety stock methods have no upper bound built into them.

The standard approach sizes a buffer from service level and lead-time variability. It will happily return a number larger than the dating ceiling, and a planning system that accepts it has just planned inventory that cannot be delivered. Cover you cannot legally ship is not cover.

Where calculated safety stock exceeds the ceiling, there is no inventory solution. The gap has to close on the responsiveness side instead:

Shorter production lead time, so the buffer needs to cover less time.
More frequent, smaller batches — subject to the minimum run problem below.
Accepting a lower service level explicitly, as a decision rather than as a surprise at intake.

Stating this as an explicit constraint is what stops the plan from quietly proposing the impossible.

Why a wide range expires its own tail

Here is the mechanism, and it is worth being precise because it looks like a forecasting failure and is not.

Take a shade range where demand is uneven — which is every shade range. Produce every shade in the same batch size, because that is what the filler's setup makes convenient. The fast shades turn quickly; the slow shades sit. At the same unit count, a slow shade represents far more weeks of cover than a fast one.

So the slow shades reach the dating ceiling first, and they do so predictably. The wider the range, the longer the slow tail, and the more of it expires. No forecast error is required. Even batches applied to uneven demand produce this outcome arithmetically.

The naive fix is smaller batches for slow shades. That runs straight into a hard floor: the filler's minimum kettle or line run. Below some quantity the manufacturer will not produce at all, and that minimum is often larger than a slow shade's entire dating-limited demand.

Postponement is the actual answer

The mechanism the industry uses is postponement — hold shared bulk or an uncoloured base, and blend or fill to shade late, close to observed demand.

What that changes structurally: the minimum run now applies to something every shade consumes rather than to each shade separately. Bulk turns at the rate of the whole range, not at the rate of its slowest member, so it does not sit against the ceiling the way a finished slow shade does. The dating clock on the finished good starts later, closer to the sale.

It is not free — it requires a fill capability close enough to demand to be useful, and it moves cost from production into operations. But it is the only structural answer to the minimum-run floor, and a plan that keeps proposing smaller batches without it is proposing something the supply chain will refuse.

The exit is a date, not a curve

The last consequence changes the shape of the exit decision entirely.

For undated product, exit logic is a curve: sell-through underperforms, markdown deepens, stock clears. Price is the lever and time is flexible.

For dated product, below a threshold the stock cannot be sold into the main channel at any price. The lever stops working. And the off-price and secondary channels that might take it have their own remaining-life requirements and their own lead times — so the window for exiting through them closes before the product actually expires.

The trigger therefore has to be computed backwards from the clearance channel's intake requirement:

Exit-decision date = expiry − the secondary channel's own required remaining life − that channel's intake and transit lead time.

Miss it and the options collapse from three to one, and the last one is disposal. Which means a dated-goods plan needs a forward-looking date flag — items projected to cross the exit-decision date at current velocity — rather than a backward-looking aging report. By the time an aging report shows a problem, the decision date has usually passed.

What to hold in the plan

Four things, none of which are lot-level:

A per-item dating ceiling, computed from shelf life and the reserved share, held as a hard cap on cover.
Safety stock bounded by that ceiling, with the shortfall named as a responsiveness gap rather than absorbed silently.
Batch sizing that reflects uneven demand, and where the minimum run blocks it, an explicit postponement decision.
A forward exit-decision date per item, derived from the clearance route rather than from sell-through.

The traceability of which physical units carry which date stays in ERP and WMS, where it belongs.

See how RetailNorthstar carries a dating ceiling as a hard cap on cover, and flags items approaching their exit-decision date.

Book a Demo →

Related resources

Merchandise Planning for Health & Beauty Brands — The vertical guide this constraint sits inside
End-of-Season Exit Strategies — Exit routes and their own lead times
Weeks of Supply Formula — The metric the ceiling caps
Safety Stock Formula — The buffer that must respect the ceiling
Weeks of Supply — Glossary — Cover as a planning concept
Replenishment — Glossary — Why smaller, more frequent runs are the responsiveness lever
Aging Inventory — Glossary — Why a backward-looking report arrives too late here