Safety Stock for a Lamp Programme: How Much to Hold, and Why It Is Not Guesswork

Every buyer has an intuition about how much stock to hold, and that intuition is usually calibrated to the wrong thing: how it feels to run out. The calculation is not emotional. Safety stock exists to cover the gap between what you sell and what you can replace, over the period it takes to replace it. That means it depends on two numbers — the variability of your demand and the length of your replenishment cycle — and it is almost always the second number that deserves the attention. Shortening the replenishment leg reduces the safety stock you need, which is the cheapest form of inventory reduction there is.

What safety stock is actually protecting

Safety stock covers the mismatch between actual demand and forecast demand during the replenishment window. If you sell more than expected while an order is in transit, the buffer absorbs it. If you sell less, the buffer is larger than it needed to be and you are carrying capital you did not need to spend. Neither outcome is a failure; both are the cost of not knowing the future precisely.

The mistake buyers make is sizing the buffer against an average. An average is not a guarantee. If your weekly demand swings between wide bounds, a buffer sized to the mean will fail roughly half the time, which is not a buffer at all. Size it against the variation you actually observe, and if you do not have that data yet, start by tracking it: the first few months of a lamp programme are the best possible time to build the history you will need later.

The two inputs, and which one you control

Replenishment lead time is not the factory's production time. It is the whole chain: the decision, the order placement, production or picking, freight, customs if applicable, receiving, and getting the stock to where it sells. In an import programme the freight and customs legs dominate, which is why the mechanics in The Import Calendar: Where Six Weeks Disappear matter so much to this calculation. In a domestic stock programme, the same leg is measured in days, and that difference is the whole argument for holding the buffer differently.

• Demand variability — how much your weekly or monthly sales move around the forecast. You influence this with promotion planning and range width, but you largely inherit it from your customers.
• Replenishment lead time — the total elapsed time from deciding to reorder to having sellable units on the shelf. You control almost all of this, and it is usually much more compressible than buyers assume.

Why the sourcing model changes the answer

Compare two programmes selling the same lamp. One buys from a factory overseas at a lower unit cost with a multi-week replenishment cycle. The other buys from inventory held in the United States at a higher unit cost with a cycle measured in days. The first needs a much larger buffer to deliver the same service level, and that buffer is capital, warehouse space and risk of obsolescence.

This is the trade-off that Stocked in the US vs Importing Direct: The Real Cost Comparison works through in detail, and the conclusion is not that one model is always right. It is that the inventory you are required to hold is part of the cost of the cheaper unit price, and it should be on the same spreadsheet as the unit price rather than treated as somebody else's problem.

Where to hold it, not just how much

A single central buffer is efficient but slow to respond; stock held at the point of sale is responsive but fragmented. Most lamp programmes end up with a hybrid: the slow-moving finishes and the deep-bulk models centralised, the fast movers and the ones tied to a specific project or property pushed forward. The right split depends on how your customers order — a hospitality operator replacing broken units in a property needs speed on a handful of models, while a retailer needs breadth across a shelf.

Replenishment discipline matters as much as the buffer size. A well-sized buffer with a sloppy reorder process underperforms a modest buffer with a reliable one, which is the argument for treating reorder triggers as a specification rather than a habit. Reading Sell-Through: Reordering Lamps Before the Shelf Goes Empty covers that side.

Three practical rules

One more rule that saves money: treat the opening order as an experiment. A mixed order covering several models at low depth tells you where the demand actually is, and the second order can then be sized properly. Committing the whole season's budget to a single deep order before you have data is the most expensive way to learn what sells. That is the reasoning behind Mixed-SKU Opening Orders: Building a First Lamp Assortment Without Overstocking.

• Size the buffer against observed variation, not against the average — and start recording the variation if you have not.
• Attack the replenishment leg before you attack the buffer size; shortening the cycle reduces the required stock directly.
• Review the buffer per model, not per range. A fast mover and a slow finish should never carry the same logic.

Frequently asked questions

How much safety stock does a lamp programme need?

Enough to cover demand variability over the full replenishment cycle, which is different for every programme. The practical approach is to measure how much your monthly sales move around your forecast, measure the true elapsed time from reorder to saleable stock, and size the buffer against the two. Programmes supplied from domestic inventory need less because the cycle is shorter, which is often worth more than the unit price difference.

Is it better to hold more stock or to accept a longer lead time?

It depends on what a stockout costs you. For a retailer, an empty shelf loses the sale and possibly the customer. For a hospitality operator, a missing lamp delays a room return or leaves a guest room incomplete. Where the cost of a stockout is high and the product is not changing, holding stock is usually cheaper than the consequence. Where the product may be discontinued or the design is seasonal, the opposite applies.

Does safety stock apply differently to floor lamps and table lamps?

Yes, mainly because of space and pack configuration. Floor lamp cartons are long and bulky, so the buffer costs more warehouse space per unit. Table lamps are usually held in pairs and turn faster, so the same capital covers more sellable units. Model the two categories separately rather than applying one blanket rule to the range.