Distribution measures: what ACV and TDP actually say

Distribution can be counted as stores, as weighted availability, or as a points measure that combines availability across items. Each answers a different question, and the choice of measure changes what a distribution story looks like.

In the independent channel, where stores vary enormously in size, the distinction matters more than in a chain-dominated read.

What is numeric distribution?

The percentage of stores in the universe that stocked the item. It treats every store equally, which is its strength and its weakness: a small store and a large one count the same.

For a channel of broadly similar stores, it is intuitive and useful. For a mixed universe, it overstates the commercial meaning of presence in small outlets.

What is weighted distribution?

The same idea weighted by store size, usually by all-commodity volume. It answers what share of the channel's total sales potential the item is available in, which is closer to the commercial question.

Weighted measures are higher than numeric ones when an item is in the larger stores and lower when it is in the smaller ones, and that gap is itself informative about where the item is listed.

What does a points measure add?

It combines availability across a brand's items, so a brand with several items in a store contributes more than a brand with one. It captures shelf presence rather than mere presence.

The consequence is that a brand can increase its points measure by adding items without gaining a single new store, which is worth remembering when a distribution gain is reported.

How do these mislead?

By being quoted without the measure named. A distribution figure that rose could mean more stores, larger stores, or more items in the same stores, and those are three different commercial events.

They also mislead across item counts. A brand that splits one item into three has increased its item-based distribution without changing what is on shelf.

What is velocity per point of distribution?

Sales divided by the distribution measure, which separates how widely an item is available from how fast it sells where it is. It is the cleanest way to tell a distribution story from a demand story.

An item with growing sales and flat velocity per point is growing on distribution; one with flat sales and rising velocity is losing distribution while performing better where it is stocked.

Why does this matter in independent retail?

Because store size varies far more than in a chain, and because gaining distribution is a store-by-store process rather than a single listing decision. A numeric measure here can move substantially with a small commercial effect, and a weighted one can barely move despite real field effort.

Reporting both is the honest approach, and the gap between them is a useful field diagnostic.

What should a distribution report include?

The measure, the universe, the period, the item basis, and store counts. Without the item basis, a brand-level distribution figure cannot be compared to its own history after a range change.

How does out-of-stock interact with this?

Distribution measures availability as observed in the data, and an item that sold nothing in a store in a period looks identical to one that was never listed. Distinguishing them requires either an observation rule or fieldwork, and that distinction is its own subject.