How Much Inventory Should You Hold? A Practical Guide for Growing DTC Brands

A DTC brand can have six months of inventory in its warehouse and still sell out of its bestselling product next week.

That's the problem with measuring stock as a single company-wide number. The business has inventory. It just doesn't have enough of the inventory customers actually want. Cash is tied up in slow-moving variants while the products driving revenue are being reordered late, expedited or left unavailable.

The right question isn't simply how much stock should we hold? It's how much of each SKU do we need, when must we reorder it, and how much cash can we afford to commit?

This guide works through those decisions using a hypothetical DTC apparel brand. All calculations, tables and charts are illustrative, not client results or industry benchmarks. The underlying reorder-point approach is consistent with guidance from Shopify and Oracle NetSuite.

1. Start with stock cover, not a company-wide inventory target

"We want to hold twelve weeks of stock" is not an inventory policy. It's an average that could be concealing two opposite problems.

A predictable core T-shirt with a seven-week replenishment cycle needs a different stock position from a short-season sweatshirt or an accessory that can be replaced in four weeks. The most useful starting point is weeks of cover at SKU level.

Weeks of cover = Sellable units on hand ÷ Expected weekly unit demand

If a product has 2,400 sellable units and normally sells 350 units per week, it has approximately 6.9 weeks of cover. That tells you when the shelf could run empty if nothing else arrives and demand remains unchanged. It doesn't establish whether the position is healthy.

Use sellable units, not stock that is damaged, quarantined or already promised to customers. Calculate cover by size and colour where customers make their choices, rather than relying on the aggregate product style. Our weeks-of-cover guide explains the metric and the traps caused by blended averages.

One more adjustment matters: sales history can understate demand if the product was out of stock. For forecasts, review comparable in-stock periods and known promotional changes. See inventory forecasting methods for DTC brands for the demand-planning layer.

2. Your lead time tells you when to place the next order

Now take that T-shirt. It sells 350 units each week. From issuing a purchase order to having replenishment ready for customers, the process takes seven weeks. That includes supplier preparation, production, inspection, freight, customs, delivery and warehouse receiving.

Seven weeks of demand is 2,450 units. If you want a further two-week buffer against delays or unexpected sales, you need another 700 units.

Reorder point = Expected demand during replenishment lead time + Safety stock

In a continuous-review model, 3,150 units is the trigger to reorder, not the quantity to purchase. The intent is to receive the shipment while approximately 700 units remain, assuming forecast demand and lead time prove correct.

Figure 1. An order raised at the 3,150-unit trigger is projected to arrive with a 700-unit buffer. Illustrated demand and lead time are fixed assumptions; actual results may vary.

This exposes a real problem with the current 2,400-unit position: at 350 units a week, stock lasts only about 6.9 weeks, but a new order takes seven weeks to arrive. With no earlier inbound stock, that creates an expected 50-unit gap before the next receipt. Placing the order now doesn't erase the shortage. The brand needs to assess an earlier shipment, stock reallocation, demand management or whether a small stockout is commercially preferable to the cost of expediting.

The formula is deliberately simple. Oracle's inventory optimisation documentation also describes how demand and lead-time variability can be used to calculate a statistical safety-stock level.

3. Safety stock should depend on the SKU, not a blanket rule

Safety stock protects against uncertainty. But uncertainty and the cost of a shortage differ by product.

• Demand volatility: a product selling roughly the same amount each week is easier to plan than one dependent on creator posts or flash promotions.

• Supplier reliability: a consistent six-week route is different from a supplier whose actual delivery varies between six and ten weeks.

• Commercial importance: a missing bestseller may lose substantial contribution, while holding too much of a low-demand colour can create markdown exposure.

• Product lifecycle: a core item can often carry forward into another season; a dated launch or seasonal colour might not.

For a small planning function, expressing a buffer as extra weeks of expected sales can be a transparent starting point. As data quality improves, a statistical model based on demand variation, lead-time variation and a chosen service level will be more defensible. Neither approach creates a guarantee against stockouts.

The relevant economic comparison is the cost of carrying additional inventory against the expected lost contribution and recovery costs when a product sells out. You can explore that latter side using Onflair's stockout cost calculator and the worked stockout-cost guide.

4. Calculate how much to buy, not just when to buy it

A reorder point answers when. The quantity to purchase depends on the ordering system and the inventory already in the pipeline.

Assume the same brand formally reviews purchase quantities every four weeks. In a simplified periodic-review, order-up-to model, the stock target must protect demand through the replenishment lead time plus the interval until the next review. The model is described in the open inventory systems textbook by Rossetti.

Order-up-to target = Weekly demand × (Lead time + Review interval) + Safety stock

Suggested order = MAX(0, Order-up-to target − Current inventory position)

Inventory position is not necessarily the same as sellable stock on the shelf. It generally considers usable on-hand stock plus eligible open purchase orders, less backorders or other outstanding commitments. Here we assume no existing inbound orders or backorders. Even when the arithmetic gives 2,150 units, the brand still needs to confirm receipt dates and address the possible shortfall before this shipment arrives.

Both the seven-week and four-week intervals matter. Using the reorder-point quantity of 3,150 as a purchasing target would ignore demand between reviews. Conversely, placing the 2,150-unit order without checking existing POs could result in buying stock twice.

Finally, a calculated quantity is a proposal. Supplier minimums, order multiples, size curves, launch plans, forecast changes and cash commitments can all change the commercially correct decision.

5. Why three SKUs need three different buying decisions

Compare three products using their current inventory positions, expected weekly demand, total lead times and illustrative buffers. For this simplified comparison, assume there are no outstanding receipts or backorders, so inventory position equals stock on hand.

Figure 2. The accessory is above its lead-time-plus-buffer reference. The other products are below it, but the seasonal sweatshirt still requires a separate end-of-season decision. Hypothetical figures.

The sweatshirt is the important exception. Suppose the season ends in eight weeks but a replacement order takes ten. Reordering could deliver fresh inventory after the market has largely moved on. The correct response may be to accept a controlled sellout rather than create obsolete stock.

A formula should identify the decision that needs attention. It should not make that decision on behalf of the commercial team.

6. A lower unit cost can still be an expensive inventory decision

Imagine a supplier offering a discount for increasing an order from 3,000 to 5,000 units. At an illustrative $18 landed cost before any discount, the extra 2,000 units represent a further $36,000 in stock commitment.

Figure 3. Before allowing for any offered price reduction, the larger order requires $36,000 more capital. Compare the actual discount with the cash, storage and markdown consequences.

The discount might still be worth taking if demand is reliable, the product has a long selling life and the additional units will sell through quickly. But if the supplier is asking the brand to buy months of speculative demand, the discount can be smaller than the cash and margin risk it creates.

Inventory purchasing has at least two separate questions: does the business need the stock? And can it fund the commitment on the actual payment dates? For the wider buying allowance, see our open-to-buy planning guide. For how deposits, balances and MOQ changes affect liquidity, see supplier payment terms and MOQs.

7. Turn the calculation into a weekly operating process

You don't need sophisticated software to start. You need reconciled sales, usable stock, a dated purchase-order schedule, trustworthy supplier lead times and a forward demand plan. A working SKU-level review should show:

1. Demand: recent in-stock sales, forecast units, promotions, wholesale commitments and expected seasonal changes.

2. Availability: sellable stock by location and variant, with any reserved or unsellable quantities identified separately.

3. Inbound: open purchase orders, quantities, confirmed milestones and expected dates at which inventory becomes sellable.

4. Risk: projected stockout date, required reorder date, excess cover and products nearing the end of their selling life.

5. Decision: buy, defer, reduce, expedite, reallocate or accept a stockout, with a named owner and an explicit cost or trade-off.

The critical step is a time-phased stock projection. A report that says "2,400 units available" tells you where you stand today. A weekly projection of expected demand and dated inbound receipts tells you whether those 2,400 units will last long enough.

For brands with several channels or teams, connect this to a repeatable sales and operations planning (S&OP) process. Purchasing, finance, commercial and operations need to agree which forecast they are funding and which stock risks they are deliberately accepting.

Keep the review simple enough to operate every week. Good planning is not an ever-expanding spreadsheet. It's an agreed cadence in which the right people make the right calls before orders become emergencies.

Frequently asked questions

How many weeks of inventory should a DTC brand hold?

There is no defensible universal target. Lead times, demand variability, ordering frequency, product lifecycle, supplier reliability and the cost of stockouts all change the answer. Calculate an appropriate policy for each important SKU or product group rather than applying one brand-wide number.

What's the difference between weeks of cover and a reorder point?

Weeks of cover measures how long current sellable stock may last at the expected demand rate. A reorder point is the stock-position threshold that triggers the next purchase in a continuous-review system. The reorder point includes expected lead-time demand and a chosen safety buffer.

Should stock that's already on order count?

Open purchase orders generally affect your inventory position, but they cannot fulfil today's customer orders. Show pipeline stock separately with realistic receipt dates. A late inbound shipment will not prevent a stockout merely because it appears in a grand total.

How often should inventory targets be reviewed?

Monitor important SKUs weekly, with exceptions reviewed more often during peak periods or major promotions. Review the underlying demand rates, lead times, safety-stock assumptions and order constraints when conditions change. Formal purchasing approvals may operate on a different schedule.

Does more safety stock always mean better service?

No. Extra stock may reduce some shortage risk, but it consumes working capital and creates carrying and obsolescence exposure. It also cannot fix inaccurate inventory records, badly distributed size curves or inbound deliveries that arrive after the expected sellout date.

The goal isn't less inventory. It's the right inventory

A business can be overstocked and understocked at the same time. It can have healthy overall inventory turns while individual bestsellers disappear from the site. And it can negotiate a lower unit cost while quietly making cash flow worse.

The solution is not a blanket cut to inventory levels. It's a repeatable link between demand, replenishment lead times, service risk and purchasing commitments.

Onflair works with scaling consumer brands to build that operating discipline across demand planning, suppliers, inventory and fulfilment. Explore fractional COO support or discuss your inventory operation.

Sources and methodology

Author: Anthony Mellor, Onflair. Published 13 November 2026. All stock quantities, costs, supplier lead times, safety-stock buffers and charts are hypothetical Onflair calculations. They are intended to illustrate different inventory decisions, not to present verified client outcomes, measured industry averages or guarantees of service level. Reorder-point and periodic-review formulas are simplified management models; a live implementation should use time-phased demand and receipt dates, actual cost definitions and documented assumptions.

• Shopify — Reorder Point Formula: A Retailer's Guide. Supports the basic lead-time demand plus safety-stock trigger.

• Oracle NetSuite — Reorder Point Defined. Clarifies that a reorder point is an order trigger, not an order quantity.

• Oracle NetSuite — Inventory Optimization Calculations. Documents demand variability, lead-time variation and statistical safety stock.

• Rossetti — Analysis of Inventory Systems, Chapter 8. Explains periodic review, inventory position and order-up-to policies.

• Shopify Help Center — Inventory fundamentals. Describes on-hand, available and committed inventory states.