Reorder when your available, sellable inventory reaches the reorder point: daily sales velocity times total lead time, plus safety stock. Because FBA lead times are long, that point is usually reached while you still appear to have weeks of stock left.
Why does my reorder point feel too high?
Because it has to cover every unit you will sell during the entire lead time, plus a buffer. On a 60 or 90 day lead time, that is a large number of units. Ordering when the shelf finally looks low means the new stock arrives weeks after you have run out.
What signals mean I should reorder early?
Rising sales velocity, a supplier or freight lane that is slipping, and an approaching seasonal peak all move the reorder point earlier. Treat them as triggers in their own right, not just inputs you update once a quarter.
Andrew Erickson is the founder of Inventory Hero. He has spent years working with Amazon FBA sellers on demand forecasting, restock planning, and the cash flow side of running a private-label brand. Inventory Hero exists because every spreadsheet-based inventory system he tried eventually broke — usually right before Q4.
You should reorder inventory the moment your available stock drops to the reorder point, which equals your daily sales velocity times your total lead time, plus safety stock. The short version: on FBA's long lead times, that trigger fires while the shelf still looks comfortable, so "when to reorder" is almost always earlier than instinct says. Below are the formula, the leading signals that move the trigger, and the mistakes that make a reorder fire too late.
Reorder when available, sellable inventory reaches the reorder point:
Reorder point = (daily sales velocity x total lead time in days) + safety stock
The logic is simple: you have to cover every unit you will sell during the entire time the new order is in transit, plus a safety stock buffer for the days demand runs hot or the shipment runs late. See the full reorder point definition and the derivation in the restock planning guide; this article is about reading the trigger in real time, which is where most timing mistakes actually happen.
A SKU sells 20 units a day, total lead time is 75 days, and safety stock is 200 units:
Input
Value
Daily sales velocity
20 units
Total lead time
75 days
Safety stock
200 units
Reorder point = (20 x 75) + 200 = 1,700 units. You order when available stock hits 1,700, even though 1,700 units feels like a full shelf. Skip that trigger and wait until you are down to a few hundred, and the replacement order lands long after you have stocked out. Check your own trigger in the reorder point calculator.
The reorder point is a moving line. These three signals push it earlier, and each is a reason to reorder before the unit count alone would tell you to.
If a SKU's sales velocity is climbing, the reorder point you calculated last month is already stale and too low. A product that went from 12 to 20 units a day needs its trigger recalculated up, or you will plan to the old, slower pace and run dry mid-climb.
Lead time is multi-leg, and any leg can stretch. A supplier quoting longer production, a port backing up, or slower Amazon check-in all lengthen total lead time, which raises the reorder point. When the lane slips, reorder earlier even if velocity is flat. See lead time variability and reorder math.
If a SKU has a seasonal ramp, the reorder that matters is the one that lands stock before demand climbs, which means ordering well ahead of the peak on the full lead time. Waiting for velocity to rise first guarantees you miss the early part of the season.
Sometimes you recalculate and find the new reorder point is already above what you have on hand, usually right after velocity jumps or a freight lane slips. You are effectively past due, and the instinct to panic-order double is the wrong one. The move is to place the normal-sized order immediately, then decide whether to expedite a portion of it, by air or as a partial shipment, to bridge the gap until the main order lands. Expediting part of one order is almost always cheaper than the stockout it prevents on a fast SKU, and far cheaper than placing a second full order that arrives after the first and leaves you overstocked once both land.
Counting reserved and in-transit units as available cover. They are real inventory, but they cannot fill an order today, so including them makes your available number look healthier than it is and pushes the trigger late.
Using a fixed reorder date instead of a reorder point. Demand does not run on a calendar. A level-based trigger adapts to velocity; a date does not.
Ignoring the lead-time legs you do not control. Freight and customs are where the surprise weeks hide. Plan total lead time, not quoted production time.
Reordering on total on-hand instead of sellable. Stranded or aged units inflate the count without being able to sell. Plan on what can actually move.
When to reorder is not a feeling about the shelf; it is the moment available, sellable stock hits the reorder point, which on FBA fires earlier than instinct suggests. Recalculate that point on live velocity and lead time, watch the leading signals that move it, and once it fires, see how to calculate reorder quantity for how much to order.