Distribution voids and out-of-stocks: telling them apart

An item that sold nothing in a store this week either was never stocked there or was stocked and unavailable. Those are completely different problems — one is a sales opportunity, the other is a supply or replenishment failure — and in transaction data they look identical.

Separating them is one of the highest-value pieces of analysis available to a brand in a channel without shelf audits.

What is the standard approach?

A rule based on history. A store that has sold the item before and has now stopped is a candidate out-of-stock; a store that has never sold it is a void. The rule needs thresholds and it needs to account for stores that genuinely delisted.

Every element of that is a judgment, which is why the rule should be written down and applied consistently rather than tuned until the answer looks reasonable.

What confounds the rule?

Slow-moving items, where a zero week is normal and says nothing. Reporting gaps, where the store did not transmit rather than not selling. And seasonal items, where absence is expected at certain times.

The reporting-gap case is the most dangerous because it produces a false out-of-stock signal that leads a field team to visit a store with full shelves.

How is the reporting problem handled?

By checking store-level completeness before applying the rule. A store with no transactions at all in a period is not reporting; a store with transactions but none of your item is a genuine signal.

That check is simple and it is the step most often omitted, which is why out-of-stock estimates from transaction data have a reputation for false positives.

What does the distinction change commercially?

Everything about the response. A void is a listing conversation with the store or the wholesaler; an out-of-stock is a replenishment or supply conversation. Sending a field team with the wrong brief wastes the visit.

It also changes the size of the prize. Voids represent unrealised distribution; out-of-stocks represent lost sales in stores that already chose to stock the item, which is usually the cheaper opportunity to recover.

How should the opportunity be sized?

For voids, by applying comparable stores' velocity to the non-stocking stores, discounted for the likelihood that non-stocking stores differ systematically. For out-of-stocks, by the store's own historical velocity over the absent period.

The second is more reliable because it uses the store's own behaviour. The first requires an assumption that stores which have not stocked would perform like those that have, and that assumption usually deserves a discount.

What is the limit of this analysis?

That it infers shelf conditions from sales. A store with stock sitting in the back room and an empty shelf sells nothing and looks like an out-of-stock, which is correct in effect and misleading about cause.

Field observation remains the only way to distinguish a supply failure from a replenishment failure, and that distinction determines who needs to fix it.

How should it be reported?

As two separate lists with different owners and different actions. A combined "opportunity" list that mixes them produces a field brief nobody can act on efficiently.