How do you forecast inventory for a DTC brand?

Start with sales velocity, your average units sold per day, then account for lead time, the days from placing an order to having it sellable, and add safety stock to cover variability. Your reorder point is velocity times lead time, plus safety stock. When on-hand stock hits that number, you reorder. It is simple arithmetic that prevents most stockouts without a forecasting tool.

What is a reorder point?

The reorder point is the inventory level at which you place your next order. The formula is sales velocity multiplied by lead time, plus safety stock. It answers the only question that matters operationally: not how much you have, but when you need to act so you do not run out before the next shipment lands.

How do you account for seasonality in a forecast?

Use the velocity from a comparable period, not a flat average. If you are forecasting for the fourth quarter, base velocity on last fourth quarter and recent trend, not the slow summer. The simple model still works; you just feed it the right velocity for the season you are forecasting, and you build headroom earlier for the peak.

Why is overstocking as bad as stocking out?

Stocking out costs you the sale and your search rank. Overstocking costs you cash tied up in product, storage fees, and, on Amazon, a worse IPI and aged-inventory surcharges. Good forecasting is not about always having plenty; it is about holding the right amount, enough to never run out, little enough to not drown in carrying cost.

Playbook Operations systems

Inventory forecasting for DTC brands, without a data team

You do not need a data scientist to forecast inventory. Here is the simple, defensible model, velocity, lead time, and safety stock, that keeps you in stock without tying up cash in overstock.

January 29, 2026 8 min read

Inventory forecasting DTC operators dread sounds like something that needs a data scientist and a model. For almost every DTC brand, this kind of ecommerce demand forecasting needs a spreadsheet and three numbers. The goal is not a perfect prediction, which is impossible; it is to never run out of a winner and never drown in cash tied up in a dud. Here is the simple, defensible method that does that, and the two failure modes it protects you from.

Inventory forecasting DTC needs only three numbers

Everything in this model starts with three inputs you already have, and together they answer how much inventory to order and when.

Sales velocity. Your average units sold per day for the product. Pull it from a recent, representative period, not a flat lifetime average.

Lead time. The number of days from placing a reorder to having that stock sellable: manufacturing, shipping, and, for FBA, the time to check in. This is the number most brands underestimate, and it is what gets them caught.

Safety stock. A buffer to cover the days when sales spike or the supplier is late. It is the cost of certainty, and a modest buffer is far cheaper than a stockout.

Forecasting is not predicting the future. It is making sure a slow supplier or a good week never costs you the sale.

The reorder point, in one formula

Put the three numbers together in one reorder point formula and you get the only operational question answered: when do I reorder?

Calculate the reorder point

Reorder point equals sales velocity multiplied by lead time, plus safety stock. If you sell 20 a day, your lead time is 30 days, and you hold 10 days of safety stock, your reorder point is 20 times 30, plus 200, which is 800 units. When on-hand stock hits 800, you order, and you arrive at the next shipment without running dry.

Use the right velocity for the period

Do not feed the formula a flat average if you are forecasting a season. For fourth-quarter planning, use last fourth quarter plus your recent trend, not the quiet summer. The model does not change; the velocity you give it does.

Correct for stockouts in your history

If you were out of stock for part of the period you are measuring, your velocity is understated, because you could not sell what you did not have. Adjust upward, or you will forecast from a number the stockout artificially suppressed and run out again.

Avoid both failure modes

Forecasting protects you from two expensive mistakes, and good forecasting respects both.

Stocking out costs you the sale and, on Amazon, your hard-won search rank, which is slow to win back. Overstocking costs you cash locked in product, storage fees, and a worse IPI and aged-inventory surcharge. The art is the middle: enough to never run out, little enough to not carry dead weight.

A forecasting routine that holds

A per-SKU sheet: velocity, lead time, safety stock, reorder point, on-hand
Velocity pulled from a representative period, stockouts corrected
Seasonal velocity used for seasonal planning
Lead time measured honestly, including check-in time
Reorder triggered when on-hand crosses the reorder point
Reviewed weekly, with action when a SKU crosses the line

This is the unglamorous core of operations-systems work, and it pairs directly with multi-channel inventory sync: the sync keeps your channels honest about what you have, and the forecast keeps you honest about when to get more.

If your stockouts and overstock keep swinging from one extreme to the other, building the forecasting and reorder discipline that ends it is exactly the kind of work a Growth Audit puts in place.