What is Sales Velocity?

Sales velocity is the rate at which a product sells, expressed as units per day or per week rather than as a total. It is the same underlying data as your unit sales, framed as a rate so that periods of different lengths can be compared.

Why the rate matters more than the total

Two things key off velocity rather than volume.

The first is rank. Sales velocity and conversion rate are what earn organic position, and both are rates. A month with a large total but a collapse in the final week reads very differently to the algorithm than a steady month with the same total.

The second is stock cover. Days of supply is sellable units divided by velocity, and it is the number that decides whether you can afford to push at all.

A worked example

Suppose an ASIN sells 300 units over 30 days.

Velocity      = 300 ÷ 30 = 10 units per day
Stock on hand = 250 sellable units
Days of cover = 250 ÷ 10 = 25 days

If a ranking push doubles velocity to 20 units per day, the same 250 units now cover roughly 12 days rather than 25. The push has quietly halved your runway, which is why our doctrine treats inventory cover as something to check before any aggression at the top of search, not after.

Choose the window deliberately

A velocity figure is only as honest as the window behind it. Measure over a period that begins after the last regime change: a deal, a stockout, a price change or a seasonal peak. Averaging across a stockout understates velocity, and averaging across a deal week overstates it. Feeding a December rate into a January plan is the classic version of this mistake.

Velocity is also a per marketplace number. The same ASIN can sell at very different rates on amazon.co.uk and amazon.de, and a blended figure hides the market that is about to run out.

How Shurq uses it

The bidding engine computes days of supply from sellable stock divided by a trailing thirty day velocity, then folds the resulting inventory zone into the bid as a multiplier. Where inventory data is unknown or has not been fed, the multiplier is neutral by design, so missing data is never treated as an emergency.

When stock is genuinely tight, the correct response is a lower bid rather than a pause. A pause removes the listing from the auction entirely and takes the conversion signal with it, while a lower bid slows spend and keeps the signal alive.