Safety Stock and Reorder Point in Foodservice Wholesale

A stockout on a high-turnover SKU is almost never an accident. It is the result of an implicit setting: orders are placed "when we see stock running low," and the threshold lives in the mind of the person placing the order. The day that person is on leave, the supplier delays by a week, or a customer doubles their order rate, the implicit threshold no longer holds. Safety stock and reorder point exist solely to document that threshold in black and white, using measurable data rather than gut feeling.

This article provides the calculation method, a complete numerical example, and—most importantly—the part that general guides omit: how minimum packaging and palletised delivery reshape the theory. In foodservice wholesale, you do not order the quantity the formula indicates. You order the next available packaging multiple, and this constraint changes the entire reasoning.

Safety Stock: What the Term Actually Means

Safety stock is the quantity you hold in addition to your expected consumption during the lead time. It does not cover normal consumption—that is covered by cycle stock. It covers only the gap between what you forecasted and what actually occurs.

This distinction is the source of most sizing errors. Safety stock is not "a month’s buffer." It is a cushion against two specific risks, and only two:

• Demand variability — a week where sales significantly exceed the average.
• Lead time variability — a delivery that arrives later than usual.

Anything not related to these two risks is not addressed by safety stock. A sustained increase in consumption is managed by revising the average consumption. A consistently late supplier is managed by revising the lead time—or the supplier. This is precisely where the decision between single sourcing and multi-sourcing arises.

Safety stock that absorbs a structural problem is no longer safety stock: it is a structural problem silently draining your cash flow every month.

The Three Metrics to Measure Before Any Formula

No formula yields useful results if the three input metrics are estimated from memory. They must be recorded, and recorded over a period representative of your business.

1. Average Consumption per Period

Use a time unit that matches your ordering cycle—weekly works for most replenishment activities. Record actual outflows over at least eight to twelve weeks. If your business is seasonal, do not mix a low and a high period in the same average: you would get a number that describes neither regime. Procurement planning based on seasonality requires separate parameters for each season.

2. Actual Replenishment Lead Time

The lead time to use is not the one stated in the catalogue: it is the time observed between the moment you decide to order and the moment the goods are available in stock. It includes internal approval time, preparation time, transport, and receipt. A receipt that takes two days to inspect and shelve is part of the lead time—on this point, checks to perform before signing the delivery note also determine when the batch becomes truly available.

3. Variability

This is the metric most often skipped, yet it is the one that sizes the buffer. Two records are enough to start: the consumption of your highest week over the observed period, and the longest lead time recorded in your last deliveries. These two extremes are more reliable than a standard deviation calculated on data too sparse to be meaningful.

The Formula and a Worked Example

Two practical formulations exist. The first covers demand variability, the second lead time variability.

• Demand risk: safety stock = (peak consumption − average consumption) × lead time
• Lead time risk: safety stock = average consumption × (peak lead time − average lead time)

When both risks are real, add the two buffers. The reorder point follows directly:

Reorder point = (average consumption × lead time) + safety stock

The following example is entirely fictional for illustration purposes. The figures do not describe any real client or SKU: they serve only to walk through the method. Let’s take a gourmet retailer selling bulk almonds.

Parameter	Value (fictional example)	Source
Average consumption	12 kg / week	Outflows recorded over 10 weeks
Peak weekly consumption	20 kg	Same record, high value
Average lead time	1 week	Observed over last 6 orders
Longest lead time observed	2 weeks	Same history, high value
Demand buffer	(20 − 12) × 1 = 8 kg	Calculation
Lead time buffer	12 × (2 − 1) = 12 kg	Calculation
Safety stock adopted	20 kg	Sum of both buffers
Reorder point	(12 × 1) + 20 = 32 kg	Calculation

Interpretation: as soon as the almond stock drops to 32 kg, the order is placed. Below this threshold, the probability of reaching zero before delivery is no longer negligible.

One point in this example is worth noting: the buffer tied to lead time (12 kg) weighs more than the one tied to demand (8 kg). This is common, and counterintuitive. Many buyers monitor sales closely and lead times loosely, yet it is often the lead time that drives the sizing.

How Packaging Reshapes the Theory

The formula yields 32 kg. In foodservice wholesale, you will not order 32 kg. You will order the next available packaging multiple, and this constraint is not a minor execution detail: it alters the decision.

At Palimex, the minimum packaging is 5 kg and delivery is palletised, with free delivery. The calculated reorder point thus becomes a trigger threshold, while the order quantity follows a different logic: available format and free delivery threshold. The choice between bucket, carton, or 25 kg bag determines your order increment, and thus the granularity with which you can adjust.

In practice, three simple rules suffice:

1. The threshold triggers, the format determines the quantity. Do not lower the reorder point because the packaging is large; accept a higher average stock.
2. The free delivery threshold is met across the entire order, not SKU by SKU. This is what makes bundling SKUs cost-effective, as detailed in our guide on free delivery in bulk orders.
3. A SKU reaching its reorder point pulls the others. When one SKU triggers, check those nearing their threshold: including them now costs less than a second shipment.

This third rule is the true hallmark of wholesale. In retail, each SKU is managed independently. In palletised procurement, SKUs are managed in waves, because transport costs are shared. The individual reorder point remains the signal; it is the order window that becomes collective.

The Other Side: The Cost of an Overly Generous Buffer

It is tempting to solve the problem by inflating safety stock. Stockouts disappear, indeed. Three costs take their place.

The first is financial: stock tied up is cash that is not working. This is the carrying cost of dormant stock, and it is quantifiable. The second is physical: storage space is not expandable, and an excessive buffer on one SKU is paid for by an insufficient buffer on another.

The third is unique to food, and it is the most costly: slow-moving stock ages. Dried fruits, candied fruits, and spices have shelf lives that do not wait for your turnover. A buffer sized by fear rather than measurement ends up as lots to clear in a hurry—and the question of what to do with a lot nearing its shelf life then arises under poor conditions, with margins already eroded.

The right sizing is therefore not the most cautious. It is the one that makes stockouts unlikely at the lowest cost, which requires accepting that they remain possible.

How Often to Review Your Parameters

A reorder point is a living parameter. Three events require recalculation, and they are easy to spot:

• A sustained change in demand — gaining or losing a regular customer, a menu change, or opening a second location.
• A change in lead time — new supplier, new logistics, or simply a series of slower deliveries than before.
• A change in packaging — switching from 25 kg bags to a smaller format, or vice versa, changes the order increment and thus the average stock.

Outside these events, a quarterly review of key SKUs is more than sufficient. There is no need to recalculate the entire catalogue: the SKUs that warrant a written reorder point are those where a stockout actually costs you. For the rest, visual monitoring remains economically rational.

Implementing the Method Without Overhauling Everything

Rollout is done by impact, not alphabetically. Start with the SKUs that meet two conditions: high turnover and a stockout that halts production or loses a sale. These rarely exceed fifteen to twenty SKUs, even in a large catalogue.

For each, record the three metrics, calculate the reorder point, and document it where the decision is made—on the shelf, in your inventory management tool, or on the stock card. A threshold that is not visible at counting time will not be followed.

Information and traceability requirements for food products are outlined by the DGCCRF, whose practical guides specify the regulatory framework for labelling and batch tracking.

After a few cycles, the benefits of the method appear beyond just avoiding stockouts: orders become predictable. You stop placing emergency orders outside free delivery thresholds, your supplier sees a regular cadence, and commercial discussions shift in nature.

Frequently Asked Questions

What is the difference between safety stock and minimum stock?

Safety stock is the calculated buffer against demand and lead time risks. Minimum stock is often a manually set threshold, without a method. The two may coincide, but only the first is justified by calculation.

Do I need a reorder point for every SKU?

No. Reserve it for high-turnover SKUs where a stockout has a real cost. For slow-moving SKUs, the cost of managing the parameter outweighs the benefit.

What if I have no outflow history?

Start by recording eight weeks of outflows for your main SKUs. In the meantime, use a cautious buffer, and replace it with a calculated one as soon as the data is available.

How do I handle highly seasonal demand?

Calculate separate parameters for each season rather than an annual average, and switch parameter sets at a fixed date before the season starts.

Does the 5 kg minimum packaging change the calculation?

It does not change the trigger threshold, but it changes the order quantity: round up to the next packaging multiple. Average stock is mechanically higher as a result.

Should payment terms be included in lead time?

No, unless financial approval is required for shipment. In that case, approval time is part of the lead time and must be counted.

How do I account for an unreliable supplier?

Use the longest lead time observed, not the stated lead time. If the gap is large and permanent, the issue is sourcing, not stock sizing.

Should safety stock be recalculated after every stockout?

A single stockout is not conclusive. Two stockouts in the same quarter, however, indicate that one of the input parameters has changed.

Can a reorder point be expressed in days rather than kilos?

Yes, provided average consumption is stable. Expressing it in quantity is still preferable, as it is the quantity that is visible at counting time.

How do I align reorder points with grouped order windows?

The reorder point signals the need; the window groups SKUs near their thresholds to meet the free delivery threshold in a single shipment.

What if a SKU hits its threshold right after delivery?

This signals that the reorder point is too low or consumption has increased. Recalculate average consumption based on recent weeks.

Do I need safety stock for a short-shelf-life SKU?

Yes, but reduced, and offset by a higher order frequency. For these SKUs, the risk of spoilage often outweighs the risk of stockout.

How do I measure if the method is working?

Track two metrics: the number of stockouts on parameterised SKUs, and the number of orders placed outside free delivery thresholds. Both should decrease.

Does the calculation apply to spices as well as dried fruits?

The method is identical. Only the parameters differ, particularly lead time and shelf-life sensitivity, which typically vary between product families.

Who in the company should own the reorder point?

The person who counts, not just the one who orders. A threshold known only to the buyer fails as soon as they are absent.

How long does it take to see results?

Allow two to three full replenishment cycles. This is the time needed for parameters to be tested in real situations.

Further Reading

• Wholesale
• Brands
• BUCKET MEYVA - EXTRA CHILI WALNUT KERNELS 3KG

Further reading

• Olives & Spreads
• Meyva
• Cycle Counting: Counting Without Halting B2B Operations
• MEYVA BUCKET - WALNUT KERNELS EXTRA CHILE 3KG