Dead stock feels quiet while stockouts feel urgent, yet the quieter problem can cost more. In retail buying, ordering too little creates stockouts, while ordering too much creates dead stock. Buyers usually fear stockouts more because an empty shelf gives an immediate signal. Dead stock remains in the stockroom until a markdown discussion arrives, often at season's end, when much of the damage is already done.
The emotional imbalance is real, but the cost often runs the other way. Working with mid-size retail chains in Spain on reorder processes, we consistently find that dead stock costs more across a full season than the stockouts buyers sought to avoid through overordering.
Why stockouts feel worse than cost
A stockout's obvious cost is the sale missed, which appears at once in the daily sales report. If you usually sell 5 units of a SKU per day and remain out of stock for 4 days, that is roughly 20 lost sales at your margin.
That calculation misses what the customer did afterward. Someone who leaves your store and orders from your website was not lost. Someone who crosses the street to buy from a competitor may or may not return. Those outcomes differ materially, yet neither appears in the stockout count.
Category also matters. In replenishment categories such as basics, consumables, and commoditized items, customers can more readily switch when you are out of stock because the product is widely available. With exclusive or distinctive merchandise, they are more likely to wait or return. A stockout in own-brand knitwear hurts less than one involving a widely distributed brand SKU carried by five other retailers in the city.
Stockout costs that stay hidden
Three stockout costs usually stay out of a P&L review:
Customer lifetime value reduction. Buyers who face a stockout during a time-sensitive purchase, gift purchase, seasonal need, or replenishment of something they rely on are measurably less likely to return. The defection rate varies by category and prior loyalty, but the direction is consistent.
Review and reputation exposure. A frustrating trip ending at an empty shelf is more likely to produce a review than a neutral visit. That review remains visible on Google and Tripadvisor for months. One "they never have stock" comment in search results can keep hurting the business long after the original stockout.
The emergency reorder premium. A stockout that demands rushed replenishment brings a freight surcharge and often weakens your bargaining position on order quantity. Expedited delivery adds to the stockout's lost revenue.
Calculating dead stock's full carrying cost
Dead stock is inventory bought at a margin it will not achieve. The clearest cost is the markdown: paying EUR 45 per unit and selling it for EUR 18 at season's end creates a EUR 27 loss on that unit, plus the original gross margin target.
Three quieter costs build at the same time:
Warehouse and stockroom occupancy. A unit left on a shelf for 14 weeks uses space that a higher-velocity SKU could take. For fashion retailers handling 200 to 500 active SKUs per season, slow movers create a real operating cost when they prevent stronger sellers from receiving prime shelf space.
Tied-up working capital. Each dead-stock unit is cash that cannot fund a reorder of something likely to sell. A buyer holding 180 units of a dead SKU at EUR 40 landed cost has EUR 7,200 tied up, earning nothing and worsening each week. For a small chain with a tight open-to-buy budget, that limits every later purchase decision in the season.
The markdown cascade effect. After choosing to discount a slow mover, you may need a deep cut to clear it before season's end. If the SKU is a distributed brand, competitors may match the reduction. The margin recovered rarely pays for the full carrying cost of holding the stock through the season.
Why each location needs a different balance
A central buying plan averaged across stores can hide location-level problems. Aggregate sell-through for a SKU may show 72 percent sold, which appears reasonable. But the Barcelona store sold out at week 6 and lost three weeks of possible sales, while the Madrid store holds 40 units that will need a 30 percent markdown at week 12.
The combined result looks acceptable. By location, both failures occurred: one store had a stockout while another accumulated dead stock. One reorder decision based on the combined figure made each problem worse.
This is why location-level inventory visibility changes the buying task. The useful question is not "how much did we sell across the chain?" It is "which stores are low on which SKUs now, and which hold more than they need?"
What data-based ordering targets
When Stockagile calculates a reorder recommendation, it works by location: how many days of cover does this store have for this SKU, based on its current sales rate and the supplier's usual lead time? If one location has 8 days of cover and another has 28, they need different reorder decisions, and one average recommendation handles them poorly.
The aim is not to eliminate dead stock and stockouts entirely. We are not saying software removes the uncertainty from buying. New collections have no sales history. Trend-sensitive categories have demand swings that models struggle to capture. A nearby competitor can alter demand in ways historical data cannot anticipate.
Location-level forecasting targets the base case: predictable depletion of established SKUs, where each store's sales history indicates with reasonable confidence when and how much to reorder. For these SKUs, cutting both dead stock and stockouts is achievable because the demand signal is strong enough to use.
The buying call data cannot replace
Season-opening buys for new collections occur before sales data exists. They remain judgment calls grounded in market knowledge, supplier relationships, trend reading, and the buyer's experience of similar products in prior seasons. No demand forecast helps with that decision.
Forecasting helps with in-season reorders. Once a SKU has a few weeks of sales history at each location, the data becomes useful. Its value depends on demand noise and on how much history the model has from comparable prior periods.
Buyers gain most from location-level forecasting when they keep first buys as judgment calls and apply data to in-season reorders. That division of labor, model for reorders and judgment for opening buys, is where the practical gain lies.
Dead stock and stockouts share one underlying gap: ordering without accurate, per-location visibility into each store's needs. The outcome depends on whether the estimate was too conservative or too aggressive. Either way, the remedy points in the same direction.