What a Reorder Point Actually Does
A reorder point is the stock level at which you place the next purchase order for an item. It exists because replenishment is never instant. The order you raise today arrives in a week, or in five weeks if it is coming by sea, and you keep selling or consuming the item the whole time you wait. If you only reorder when the shelf is empty, you spend the entire lead time out of stock.
The reorder point answers one question: when should I order? The separate question of how much should I order? belongs to order quantity. Companies often get one right and the other wrong. They carefully calculate an economic order quantity, then release orders whenever a buyer happens to notice the shelf looks thin. Or they trigger orders at a sensible level but in quantities so small that they are back below the trigger a few days after each delivery.
This guide walks through the formula, a full worked example, three ways to size safety stock, how service level targets change the numbers, the mistakes that wreck a correct formula, and what an ERP system changes when you have hundreds or thousands of items to manage.
Why Reorder Point Inventory Management Matters
Reorder point inventory management sounds like a warehouse detail. In practice it touches most of the business:
- Revenue. A stockout on a fast-moving item is a lost order, and in B2B distribution it is often a lost share of the customer. Trade buyers in the Kingdom typically keep two or three suppliers on rotation. Miss twice and the volume shifts.
- Production. In a factory, one missing raw material can idle a whole line. The cost is not the material itself but the labour, machine time and late delivery penalties that follow.
- Cash. Set the trigger too high and you are always ordering early and carrying more than you need. That stock is cash that could be paying suppliers or funding growth.
- Purchasing discipline. Without a clear trigger, buying becomes reactive. Reactive buying means rush orders, air freight, and accepting whichever supplier can deliver fastest at whatever price they name.
The reorder point is where two goals collide: never running out, and holding as little as possible. A well-set reorder point does not make that tension go away. It turns the trade-off into a deliberate decision you can explain and adjust.
The Formula and Its Three Inputs
The standard formula is:
Reorder point = (average daily demand × lead time in days) + safety stock
The first part is lead time demand: what you expect to sell while the order is on its way. The second part is the buffer against everything that does not go to plan. Each input hides decisions that change the answer.
Average daily demand
This is how many units leave stock on an average working day, whether sold, issued to production or sent to a project site. Three choices matter:
- The look-back window. A twelve-month average blends peak and off-peak seasons together. Last week's average is distorted by any unusual order. A rolling 8 to 13 weeks, with seasonality handled separately, usually works better.
- Working days, not calendar days. If the warehouse ships six days a week, divide by working days. Dividing by seven understates daily demand by about 14%.
- Outliers. A one-off tender or a project order that will not repeat should be stripped out and planned as a special requirement, not baked into the average.
Lead time
Lead time runs from the moment you decide to buy to the moment the item is available to sell or issue. It is longer than the transit time the supplier quotes. For an imported item it typically includes:
- Raising and approving the purchase order internally, which can take two or three days if several sign-offs are required.
- Supplier confirmation and order preparation.
- Transit, which is days for a local supplier and weeks for ocean freight.
- Port handling, customs clearance and the truck from the port to your warehouse.
- Receiving, inspection and putting the stock into the system.
Using the supplier's transit time alone is one of the most common reasons a reorder point comes out several days short, which then shows up as "unexplained" stockouts even though the order went out on time.
Safety stock
Safety stock covers the gap between expected and actual: a week where demand runs hot, or a container that sits at the port for an extra ten days. Its size depends on how much demand varies, how much lead time varies, and what service level you want. Because it involves the most judgment, it gets its own section below. For a deeper treatment, see our guide to safety stock.
A Worked Example
Take a hypothetical auto parts distributor in Riyadh selling a popular oil filter. Thirteen weeks of history show:
- Average demand of 25 units per working day, with a standard deviation of 8 units.
- The filter is imported through Jeddah. Lead time from PO approval to stock on the shelf averages 20 working days, with a standard deviation of 5 days. The fastest order took 14 days and the slowest took 32.
- Unit cost is SAR 35.
- Management wants the item available in 95% of replenishment cycles.
Step 1, lead time demand: 25 × 20 = 500 units.
Step 2, safety stock (demand variation only): service factor × demand standard deviation × square root of lead time. The factor for 95% is 1.65 and the square root of 20 is about 4.47. So 1.65 × 8 × 4.47 = 59 units.
Step 3, reorder point: 500 + 59 = 559 units.
That assumes the lead time is always 20 days. It is not. It ranged from 14 to 32. That is where the real risk sits.
Three Ways to Size Safety Stock
Method 1: demand variation only
This is what we just did: 59 units. It is reasonable for a reliable local supplier who delivers on almost the same day every time.
Method 2: demand and lead time variation combined
The formula becomes: service factor × square root of [(lead time × demand std dev²) + (average demand² × lead time std dev²)].
- Lead time × demand variance: 20 × 64 = 1,280
- Average demand² × lead time variance: 625 × 25 = 15,625
- Total 16,905, square root about 130
- Safety stock: 1.65 × 130 = 215 units
The reorder point becomes 500 + 215 = 715 units. Look at the two terms inside the square root: lead time variability contributes over 90% of the total. At SAR 35 a unit, the extra buffer is worth about SAR 5,500 in this one item. Multiply that across a few hundred imported SKUs and supplier reliability turns out to be a balance sheet issue, not just a purchasing one. Tightening lead times often cuts inventory more than better forecasting does. Our guide on lead time management covers how to measure and improve this.
Method 3: max minus average
A popular shortcut that needs no statistics: (maximum daily demand × maximum lead time) minus (average daily demand × average lead time). If the highest day on record was 40 units and the slowest delivery 32 days: 1,280 − 500 = 780 units of safety stock, for a reorder point of 1,280.
It is easy to explain and very conservative, because it assumes the worst demand and the worst delay always arrive together. Use it as a starting point when you have little data, or for a genuinely critical item, but expect to carry far more stock than you need.
| Method | Safety stock | Reorder point | Best suited to |
|---|---|---|---|
| Demand variation only | 59 | 559 | Reliable local suppliers |
| Demand and lead time variation | 215 | 715 | Imports, erratic suppliers |
| Max minus average | 780 | 1,280 | Thin data, critical items |
Service Level: How Much Protection Is Enough?
Cycle service level is the probability of not running out during a replenishment cycle. Raising it increases safety stock, and the cost climbs faster as you approach 100%. Using the oil filter example with demand variation only:
| Service level | Factor | Safety stock (units) |
|---|---|---|
| 90% | 1.28 | 46 |
| 95% | 1.65 | 59 |
| 97% | 1.88 | 67 |
| 98% | 2.05 | 73 |
| 99% | 2.33 | 83 |
Going from 90% to 95% adds 13 units for five points of protection. Going from 98% to 99% adds 10 units for a single point. Applying 99% to every SKU is an expensive habit. A more sensible policy ties the target to the item's importance:
- High-value A items and anything whose absence stops production: 97 to 99%.
- Mid-range items: around 95%.
- Slow movers and items with ready substitutes: 90% or lower.
Compare Against Inventory Position, Not Shelf Stock
This mistake catches people who calculate the formula perfectly. You should not compare the reorder point with what is physically on the shelf. You should compare it with inventory position:
On hand + on order − allocated to customer orders not yet shipped
Say the warehouse holds 420 filters, a previous PO for 600 is in transit, and 90 are allocated to a customer collecting tomorrow. Inventory position is 420 + 600 − 90 = 930. On-hand stock is below the reorder point of 559, but you should not order, because the shipment already on its way covers the need. Trigger on shelf stock alone and you place a duplicate order, then find yourself with twice the stock you planned two weeks later.
Reorder Point and Order Quantity Work Together
Once you know when to order, you still need to decide how much. Common approaches:
- Economic order quantity, which balances ordering cost against holding cost. See our guide to EOQ.
- Supplier minimums, such as a full pallet, a full truck or a minimum invoice value.
- Order-up-to, where you order whatever brings inventory position back to a set maximum.
The two settings interact. If the order quantity is small relative to lead time demand, each delivery barely lifts you above the trigger and you end up placing order after order. As a rough rule, the order quantity should be at least close to lead time demand.
Continuous Review vs Periodic Review
The formula assumes continuous review: you watch stock all the time and order the moment it hits the trigger. That works when every transaction is posted in real time. Many businesses instead review weekly, or order from a given supplier on a fixed day when the sales rep visits.
With periodic review, an item can cross the trigger the day after a review and wait almost a full cycle before anyone orders it. Stock must therefore cover lead time plus the review period. In our example, a weekly review (six working days) means covering 25 × (20 + 6) = 650 units of demand before safety stock, and safety stock rises too, because the window of uncertainty is longer.
| Aspect | Continuous review | Periodic review |
|---|---|---|
| When you order | When stock hits the trigger | On a fixed schedule |
| Order size | Usually fixed | Varies with stock level |
| Safety stock | Lower | Higher, covers review gap |
| Needs | Real-time transaction posting | Regular count or report |
| Fits | Fast, important items | Consolidating orders per supplier |
When to Recalculate
A reorder point is not a set-and-forget number. Revisit it when:
- Seasons shift. Food and beverage demand in the Kingdom climbs ahead of Ramadan, AC parts ahead of summer, stationery ahead of the school year. Use last year's in-season demand rather than the annual average, and adjust at least one full lead time before the peak starts.
- You change supplier. A new supplier comes with a different lead time and an unknown track record. Hold extra safety stock until you have a few deliveries of real data.
- Freight conditions change. Port congestion or rerouted shipping lanes stretch lead times and their variability. Track actual receipt dates, not quoted ones.
- A product is being replaced. Demand for the old item tapers. If its reorder point stays where it was, you build up stock that may never sell.
Mistakes That Break a Correct Formula
- One rule for everything. "Reorder at two weeks of stock" is simple but ignores that each item has its own supplier, lead time and volatility.
- Inaccurate stock records. If system balances do not match the shelf, any formula produces the wrong answer. Regular cycle counting comes before reorder point tuning.
- Measuring demand from fulfilled sales only. When an item is out of stock, recorded sales drop because you could not sell, not because customers stopped asking. Use those figures and you lower the trigger, which causes more stockouts next cycle. Capture lost or declined orders too.
- Unit of measure confusion. Buying in cartons and selling in pieces, or buying by the tonne and issuing by the kilo, quietly corrupts every calculation if conversions are not set up properly.
- Undocumented assumptions. When finance asks why one SKU carries so much stock, the answer should already be written down: 98% service level because it is critical, 45-day lead time because it is imported.
Most stockouts do not come from a wrong formula. They come from bad data going into it, or a correct trigger that nobody acted on.
What an ERP System Changes
Ten items in a spreadsheet is manageable. A thousand items across three warehouses, with open POs, customer allocations and different suppliers, is not: the spreadsheet is out of date before it is finished. An ERP changes the mechanics:
- Live balances. Every sales invoice, issue and receipt updates stock immediately, so you work from today's position, not last month's count.
- Full inventory position. The system knows what is on order and what is allocated, so the trigger compares against real availability.
- History on tap. Average demand and actual supplier lead times come from posted transactions instead of guesses.
- Inventory linked to purchasing. Hitting the trigger leads to a purchase requisition that goes through the normal approval flow.
- Multiple units of measure. Buy in cartons, sell in pieces, without manual conversions.
In Tranquil, inventory management runs alongside purchasing in one cloud ERP, with central inventory and multiple units of measure, so balances and open orders sit in the same place when the buying decision is made. If you want to review how your items are set up, talk to our team.
Frequently Asked Questions
What is the difference between reorder point and safety stock?
Safety stock is part of the reorder point. The reorder point is expected lead time demand plus safety stock. With zero safety stock, the reorder point equals lead time demand and, with symmetric demand, you would run out in roughly half of all cycles.
Should the reorder point be in days or units?
The trigger itself is in units, because it is compared with stock balances. It is still worth expressing it in days of cover for discussion: 559 units at 25 a day is about 22 days of sales.
How often should reorder points be updated?
Monthly or quarterly for important, fast-moving items, plus whenever seasons, suppliers or freight conditions change. Slow movers can usually be reviewed twice a year.
How do I set a reorder point for a new item with no history?
Borrow demand from a similar item or use the sales forecast, start with higher than usual safety stock, and recalculate after two or three months of real movement.
Does a reorder point work for made-to-order items?
No. If an item is only bought against a specific customer order, purchasing is driven by that order, not by a stock trigger. Reorder points are for items you keep in stock.