Sales velocity in CPG, explained

What sales velocity is

Sales velocity is how fast a product sells once you correct for how many stores carry it, usually stated as units per store per week. When I owned weekly SPINS reporting at a natural-products brand, this was the first number I pulled and the only one I trusted on its own. A SKU doing 120,000 units a quarter looks like a winner until you learn it sits in 4,000 Kroger doors, at which point it's a quiet underperformer hiding behind a big distribution footprint.

That's the whole reason velocity exists. Raw dollar sales and raw unit sales conflate two completely different things: how many stores sell the item, and how hard it sells in each one. Velocity strips out the first so you can see the second. If you searched "velocity cpg" because a buyer quoted your numbers back at you and they didn't match your gut, this is the metric you were both circling.

How velocity is measured

There are two common forms, and analysts use both depending on who's asking. The first is units per store per week, which most people shorten to units/store/week. You take total units, divide by the number of stores selling the item (not the number that exist, the number actually scanning it), then divide by the number of weeks. The second is dollar velocity per points of distribution, written as $ per $MM ACV per week. That one normalizes by ACV-weighted distribution instead of a raw store count, which matters because a Whole Foods door moves far more volume than a single rural independent.

Here's the trap. Raw sales reward distribution gains and punish nothing, so a brand can grow total dollars 30% in a year purely by adding doors while velocity in each door quietly slides. That slide is the early warning that the product isn't pulling its weight on shelf, and it shows up in velocity months before it shows up in a category review.

Reading velocity: two brands, same category

Take two refrigerated-juice brands in the natural channel, both reporting to a category buyer in the same review. Brand A is the incumbent with broad distribution. Brand B is the challenger, narrow but sharp.

Brand A wins on every raw number: three times the units, three times the dollars. But velocity tells the opposite story. Brand B moves 26.9 units/store/week against Brand A's 20.0, which means it sells 35% harder in each door it occupies. Run the math: 312,000 / 1,200 / 13 = 20.0, and 98,000 / 280 / 13 = 26.9. Same category, same shelf, and the smaller brand is the one a buyer should expand, because it has earned every door and is starved for more.

This is exactly the read a buyer wants from you. They have finite shelf. Velocity tells them which item will reward an extra facing and which one is already coasting on the doors it has.

Why velocity is the analyst's anchor metric

Velocity is the cleanest single signal of product-market fit at shelf, and it feeds nearly every decision downstream. Distribution decisions hang on it: a high-velocity SKU starved for doors is the easiest expansion case you will ever build, while a low-velocity SKU with broad distribution is a SKU-rationalization candidate dressed up as a success. Promotion planning hangs on it too, because lift only means something measured against a steady base.

That base is the catch. A single velocity number blends promoted and non-promoted weeks into one average, which can flatter a SKU that only moves when it's on deal. To separate the two you need to split the steady baseline demand from the incremental units a promotion drives. A SKU running 30 units/store/week sounds healthy until you learn 22 of those come from a deep TPR that ran 8 of the 13 weeks. Strip the promo and the real base is closer to 12, and the expansion case you were about to make falls apart.

If you want the full decision logic for when velocity, share, and TDP each take priority, the velocity / share / TDP decision tree walks through it.

Where Scout fits

Velocity math is simple arithmetic, but doing it cleanly across hundreds of SKUs, splitting promoted from baseline weeks, and holding the store counts straight is where the weekly grind lives. Scout connects your SPINS or retailer data and computes velocity by SKU, by retailer, and by promoted versus base period, so the number you bring to a category review is already deconfounded. It measures and analyzes the rate; it doesn't decide your assortment for you. That call is still yours.

The short version

• Sales velocity is sales rate corrected for distribution, usually units per store per week or $ per $MM ACV per week.
• Raw dollars and units reward adding doors and hide a falling per-store rate. Velocity isolates how hard a product actually sells on shelf.
• A blended velocity number mixes promoted and base weeks. Split baseline from incremental before you make a distribution or promo call, or the average will mislead you.