Fast-Moving vs. Slow-Moving Inventory

The same reorder rule applied to both ends of a catalog is how a store ends up out of stock on the candle that funds the business while sitting on three years of wick trimmers. Both outcomes come from one policy doing its job correctly on products it was never sized for.

Fast movers and slow movers do not need a better version of the same plan. They need opposite plans, run at the same time, out of one purchase-order budget. This post is about running both.

What separates the two

Velocity is units sold per unit of time, measured against how much of the SKU you are holding. Turnover, sell-through, and days of inventory all describe it from slightly different angles, and all of them are relative to your own catalog rather than to a published benchmark. A SKU selling 30 a week is fast in one store and unremarkable in another.

Two things velocity is not. It is not revenue: a cheap accessory can be your fastest-moving product and still be a rounding error on the income statement. And it is not margin: your slowest mover can be the most profitable thing you sell per unit. That is exactly why velocity is one axis of a grading and not the whole of it, and why crossing it with revenue contribution in an ABC-XYZ analysis tells you more than either measure on its own.

Where the line actually sits for your store, and which metrics flag a SKU that has crossed it, is covered in how to identify slow-moving inventory. This post assumes you already know which SKUs are which.

Opposite failure modes

A fast mover fails by running out. The cost is the sale you didn't make, plus a second cost most stores never account for: demand you never recorded cannot be forecast. A week at zero looks identical in your sales history to a week of no interest, so a stockout quietly teaches your own forecast that the product is slower than it is, and the next order comes in smaller.

A slow mover fails by never running out. The cost is capital committed to units that are sitting still, plus storage, plus the slow erosion of obsolescence, damage, and going out of season. Nothing about this failure produces an event. There is no date, no unhappy customer, and no line item.

That asymmetry explains most of the bad decisions in this area. Stores over-correct toward availability because the availability failure is the one that gets shouted at them, and the money that pays for the over-correction comes out of a corner nobody is watching.

How to plan a fast mover

Fast movers are the case the standard machinery was built for, because they generate enough sales data to make the arithmetic meaningful. Four things matter.

Keep the reorder point current. Not the formula, the inputs. A trigger level calculated from three-month-old velocity is describing a store that no longer exists, and on a fast mover the drift compounds quickly because the SKU burns through its buffer in days rather than months.

Size the buffer to this SKU's own variability. This is where safety stock genuinely earns its cost: high velocity means the buffer turns over rather than sits, so you are renting protection rather than buying a warehouse. The safety stock guide covers the sizing.

Working the site's usual example: "Cedar & Fig, 250g" sells 5 units a day on a 12-day lead time. Lead-time demand is 60 units, and a 30-unit buffer, six days of cover, puts the trigger at 90.

Order more often, in smaller quantities. Frequency is a real lever here and an underused one. Shorter gaps between orders mean less cash committed at any moment and a faster reaction to a change in demand, paid for in freight and admin. On a product that sells reliably, that trade is usually worth taking.

Treat lead time as the risk it is. On a fast mover, every extra day of supplier lead time is another 5 units of stock you have to carry to stand still. A supplier who slips from 12 days to 18 has changed your trigger level by 30 units without telling you.

How to plan a slow mover

The instinct is to run the same process with smaller numbers. That is the mistake. On a SKU selling a handful of units a month, average daily sales is not a stable quantity, and a reorder point built from it inherits all of that instability while looking exactly as authoritative as the fast mover's.

Buy to a coverage period. Decide how many months of cover you are willing to fund, and order to that. It is a cruder rule and a more honest one, because it makes the decision visible as a decision rather than hiding it inside a formula that had no real inputs.

Expect the supplier's minimum to be the binding constraint. On slow movers the minimum order quantity, or the case pack, is usually larger than your coverage period calls for. That is not a rounding problem, it is the whole decision: if the smallest order the supplier will accept is eighteen months of stock, the real question is whether to keep the SKU rather than how much to buy.

Be sceptical about safety stock. A buffer on a slow mover is an expensive insurance premium against a cheap event. The stock sits for months, and the customer who wanted it will often accept a short wait, particularly on a niche product they came to you specifically for. Accepting a longer replenishment window, or taking the order as a backorder, is frequently the better trade than funding a buffer that will still be there next year.

Know where the exit is. Slow and dead are points on the same spectrum, not different categories, and the difference is whether the product still sells at a normal price. What is dead stock covers where that boundary falls and what changes once a SKU has crossed it.

Splitting one budget between them

Fast and slow movers do not compete for shelf space. They compete for money, and that competition is usually invisible because purchase orders get raised supplier by supplier rather than against a total.

Every unit of slow stock in an order is a unit of fast stock you did not buy. The clean way to make that trade visible is to fund it in order: cover your fast movers to their reorder points first, then spend what is left on the rest. Anything that cannot be funded after that is not a budgeting problem to be solved with a bigger order, it is a signal about the SKU.

The other half of the budget is the money already sitting in slow stock. That capital is not gone, it is illiquid, and converting it back is a separate exercise from planning: selling excess inventory without losing margin covers the ways to do it, ordered from least to most drastic. Stores tend to reach for the credit line before they reach for the stockroom, which is the more expensive of the two options.

When a fast mover slows down

This is the transition that costs the most, because both policies are correct and the store is running the wrong one. Parameters set for the fast regime keep firing: the reorder point still triggers, the order quantity is still sized for the old velocity, and each cycle adds another batch on top of stock that is no longer selling through.

The signal is not a bad month. It is sell-through flattening across consecutive cycles while the order size stays the same, with no seasonal or promotional explanation available. One weak month on a fast mover is noise; three cycles of falling sell-through with unchanged buying is a regime change.

The order of operations matters. Cut the order quantity first, then re-check the buffer, and only then consider price. Discounting a SKU that is still selling at full price, just more slowly, gives away margin to solve a problem that a smaller purchase order would have solved for free. Discounting belongs at the end of the sequence, once the stock is genuinely surplus rather than merely ahead of demand.

Catching that transition is the part that is hard to do by hand, because it requires noticing the absence of something across several cycles on every SKU at once. StockCue recomputes velocity and reorder points from live sales rather than leaving them where they were last set, and flags overstock and dead stock as it develops. Forecasting runs on every plan including Free, which covers 50 SKUs; purchase orders and receiving start at Starter.

Frequently Asked Questions

What counts as a fast-moving product?

Velocity is relative to your own catalog, not to an industry number. A fast mover is a SKU whose turnover and sell-through sit well above your catalog's own average, and whose days of inventory on hand is short compared with how often you place orders. The same units-per-week figure can be fast in one store and slow in another, which is why importing someone else's cutoff produces a grading that does not match the business it is describing.

How should slow-moving inventory be reordered?

Buy to a coverage period rather than to a formula. On a SKU that sells a handful of units a month, average daily sales is an unstable number and a reorder point built from it will fire at close to the wrong moment. Decide how many months of cover you are willing to hold, check that against the supplier's minimum order quantity, and if the minimum forces you to hold far more than that, the real decision is whether to keep stocking the product at all rather than how much to order.

Should slow movers have safety stock at all?

Often very little, and sometimes none. Safety stock is priced insurance: you pay in cash and shelf space to avoid a stockout. On a slow mover the premium is high, because the buffer sits for months, and the payout is low, because few customers are waiting and many will accept a short wait. Computing a statistical buffer from four sales in a quarter is arithmetic performed on noise. Accepting a longer replenishment wait, or taking the order as a backorder, is usually the better trade.

What do you do when a bestseller starts slowing down?

Cut the order quantity before you cut the price. The damage in this transition comes from planning parameters that were set for the fast regime and never moved, so each cycle reorders at the old volume against demand that has already fallen. Watch sell-through across consecutive cycles rather than a single month, and when it flattens with no seasonal or promotional explanation, reduce the next order and re-check the buffer. Discounting is a later step, and a more expensive one.