Reorder Point Calculator
Four methods, one calculator. Find the stock level that should trigger your next purchase order, and see how far the answer moves once you allow for a supplier who is sometimes late.
Calculate Your Reorder Point
Pick a method and the fields adjust. Everything runs in your browser: no sign-up, no server call, nothing you type leaves the page.
Formula applied
Fill the fields to see the formula with your numbers substituted in.
What this means
Enter your figures to see when to order and how much cover the trigger level buys.
| Method | Safety stock | Reorder point | Days of cover | Buffer value |
|---|---|---|---|---|
| 1. Basic | — | — | — | — |
| 2. With safety stock | — | — | — | — |
| 3. Demand variability | — | — | — | — |
| 4. Demand and lead time | — | — | — | — |
Buffer value uses the optional unit cost field. Rows missing an input show a dash.
Which of the Three Calculations Do You Actually Need?
These three get used interchangeably and are not the same number. Each answers a different question, and a working replenishment rule needs all three.
| The question | The calculation | What it gives you | What it needs |
|---|---|---|---|
| When do I place the next order? | Reorder point (this page) | The stock level that triggers a purchase order. | Daily demand, lead time, safety stock. |
| How much do I order each time? | Economic order quantity | The order size balancing ordering cost against holding cost. | Annual demand, cost per order, annual holding cost per unit. |
| How big should the buffer be? | Safety stock | The cushion absorbing variability. An input to the reorder point. | Demand deviation, lead-time deviation, service level. |
How they fit together
Work through them in order: buffer, then trigger level, then order size. Change one and the other two are worth re-checking.
Four Reorder Point Formulas, and Where Each One Breaks
Most pages publish the first formula and stop. It is the right starting point and the wrong finishing point: it carries no protection against a busy week or a delayed vessel.
| Method | Formula | What it needs | Use it when | Where it fails |
|---|---|---|---|---|
| 1. Basic | ROP = Average daily demand × Lead time (days) | Two numbers. No history, no statistics. | A first pass, a new SKU, or a same-city transfer. | Zero buffer, so it runs short in roughly half of all cycles. |
| 2. With safety stock | ROP = (Average daily demand × Lead time) + Safety stock | The basic inputs plus a buffer you already hold. | Your planning system or 3PL already gave you a buffer number. | Only as good as the buffer you feed it. A number picked by feel carries no service level. |
| 3. Demand variability | Safety stock = Z × σ(demand) × √(Lead time) | Daily demand deviation and a target service level. | The supplier is reliable and arrival dates barely move. | Assumes lead time is constant. On an import lane that is usually wrong by a wide margin. |
| 4. Demand and lead-time variability | Safety stock = Z × √( LT × σ²(demand) + D² × σ²(LT) ) | Everything above plus the deviation of actual lead time. | Cross-border sourcing, customs, port congestion, marketplace receiving queues. | Needs real receipt dates. Guessing the lead-time deviation low gives a comfortable wrong answer. |
Why method 4 is the one that matters on an import lane
Under the square root, lead-time variance is multiplied by daily demand squared, so on a fast mover that term dominates. Set the lead-time deviation to zero and method 4 collapses back to method 3 exactly.
Service Level to Z Score
Methods 3 and 4 need a Z score: the number of standard deviations of cover that delivers your chosen cycle service level, read one-tailed from the standard normal distribution.
| Cycle service level | Z score | Expected shortfall | Typical use |
|---|---|---|---|
| 80% | 0.84 | 2 cycles in 10 | Long-tail items customers will substitute. |
| 85% | 1.04 | 1.5 cycles in 10 | Slow movers you stock but will not protect. |
| 90% | 1.28 | 1 cycle in 10 | Steady B-class lines with a supplier who can expedite. |
| 95% | 1.65 | 1 cycle in 20 | The default for A-class and hero SKUs. |
| 97.5% | 1.96 | 1 cycle in 40 | High-velocity marketplace listings where ranking suffers. |
| 98% | 2.05 | 1 cycle in 50 | Top revenue lines and retailer fill-rate agreements. |
| 99% | 2.33 | 1 cycle in 100 | Stockouts here trigger a contractual penalty. |
| 99.9% | 3.09 | 1 cycle in 1,000 | Effectively never short. Spend on lead-time reliability instead. |
Values are the one-tailed inverse normal, rounded to two decimals. At 95% the exact figure is 1.6449; the long-standing convention rounds it up to 1.65, which this page uses.
Cycle service level is the probability of not running short during one cycle. It is not fill rate, which measures the share of unit demand served and is normally higher for the same buffer.
A reorder point on averages is short half the time
Method 1 covers demand during an average lead time, and demand exceeds its average in roughly half of all cycles. A zero-buffer reorder point is therefore a coin flip. Nothing is broken when it stocks out; the Z score is what buys those odds down.
Lead Time Is Not the Number Your Supplier Quoted
A supplier quoting fourteen days means fourteen days to make the goods. The formula wants something longer: elapsed time until the stock is sellable. Break it into steps and add them up.
| Step | Days in this example | Who controls it | What actually happens |
|---|---|---|---|
| Supplier production | 14 | Supplier | Starts when the order is confirmed and the deposit clears, not when you email the PO. |
| Booking and cargo ready to departure | 4 | Forwarder | Space booking, trucking to port, cut-off and rollover risk. |
| Port-to-port transit | 6 | Carrier | Shanghai to Long Beach by sea. Air compresses it; transhipment stretches it. |
| Import customs clearance | 2 | Broker | Document check and release. A classification or valuation query adds days. |
| Receiving and putaway | 6 | Warehouse or marketplace | Appointment slot, unload, count, label, stock live. The most commonly forgotten step. |
| Total lead time | 32 | You | This is the figure the reorder point formula wants. Sum your own steps and enter the total above. |
Day counts here are the example lane below, not published benchmarks. Measure your own from purchase order date to sellable stock.
Receiving is the forgotten week
6 days
Dock slot, unload, count, label, stock live. Invisible on the supplier invoice, visible in your stockout report.
Marketplace inbound queues
Your inbound sits behind everyone else's. Peak weeks lengthen it and you control none of it.
Customs is variable, not slow
Most entries clear fast. The risk is an occasional classification query, so it belongs in the deviation as well as the average.
Measure receipts, not quotes
Use actual dates from your last ten or more receipts. A quoted lead time is a target; receiving records are evidence.
An Imported Skincare SKU Selling Into the US
A beauty brand sells one hero serum through its own store and a regional marketplace, replenished by sea from a contract manufacturer near Seoul. Twelve months of history gives these figures.
Average daily demand
40 units
US only, across all trading days.
Total lead time
32 days
The five steps above, purchase order to sellable stock.
Demand deviation
11 units
Daily standard deviation over the same twelve months.
Lead-time deviation
7 days
Across 14 receipts. Sailings and receiving queues drive it.
It holds 400 units as an informal buffer and orders 1,600 at a time at $9 landed. At a 95% service level, each method returns this on identical data.
| Method | Substituted | Safety stock | Reorder point | Days of cover | Reading |
|---|---|---|---|---|---|
| 1. Basic | 40 × 32 | 0 units | 1,280 units | 32.0 | No buffer. Half of all cycles land short before the container arrives. |
| 2. With safety stock | (40 × 32) + 400 | 400 units | 1,680 units | 42.0 | A buffer chosen by feel. Sensible in size, but with no stated service level. |
| 3. Demand variability | 1.65 × 11 × √32 | 103 units | 1,383 units | 34.6 | Two and a half days of cover. It assumes the ship is never late; on this lane it often is. |
| 4. Demand and lead time | 1.65 × √(32 × 11² + 40² × 7²) | 473 units | 1,753 units | 43.8 | The realistic answer here. Nearly five times the buffer of method 3 on identical data. |
Lead time drives 95% of the variance under the square root in method 4. The order-up-to level there is 3,353 units, or 1,753 plus the 1,600-unit order: 84 days of cover on arrival.
The 400-unit informal buffer sits between the demand-only and full answers. Buffers set by feel are rarely absurd; they just carry no stated probability of avoiding a stockout.
What Changes During 11.11, Black Friday and Ramadan
A reorder point built on a twelve-month average is wrong in both directions during a peak. Demand rises, lead time stretches, and the two compound rather than cancel out.
Demand is not random here
A campaign spike is planned, not statistical variation. Blending it into a standard deviation describes neither well.
Do instead: plan peak cover from the campaign forecast and keep the statistical buffer for baseline weeks.
Lead time stretches too
Space tightens and receiving queues lengthen exactly when volume peaks. Both terms move against you at once.
Do instead: raise the lead-time figure for orders landing in the peak window, not just demand.
Work back from the cut-off
A reorder point is a level; a peak is a date. If the last order that can land in time is eight weeks out, the level stops binding.
Do instead: set a calendar deadline alongside the level and order to the forecast.
Ramadan moves each year
It shifts roughly eleven days earlier each year, so last year's dates do not transfer.
Do instead: anchor the peak window to the dates for the coming year, then diarise a reset once it passes.
Review the Reorder Point on a Cadence, Not Once
A reorder point is a snapshot of demand and a lane at one moment, and both drift. The failure mode is a right formula run on figures from fourteen months ago.
Review triggers and cadence
- Recalculate monthly for A-class items and quarterly for the rest, per SKU rather than per category.
- Recalculate immediately after a supplier change, a lane change, or a move to a new fulfilment centre.
- Re-measure lead time whenever a receipt lands well outside the usual spread, and after any change to order quantity.
- Exclude promotional and launch weeks from the baseline before measuring deviation.
- On a periodic-review model, add the review interval to lead time before calculating.
- Check the unit of measure end to end, and sanity-check days of cover. Four months of cover means the inputs deserve a second look.
- Strip stockout days out of the demand history. Recorded as zero, they shrink the level every time it fails.
- Compare the trigger against stock on hand plus stock already on order, or you double-order every cycle.
- Name an owner. Buffers and trigger levels drift quietly when nobody is accountable for reviewing them.
Planning and Cost Tools That Pair With This One
These free tools cover the buffer inside it, the size of the order, and what holding the result costs across a year.
Safety Stock Calculator
Size the buffer that feeds this formula, with three methods and a service level cost curve.
EOQ Calculator
Work out how much to order once the trigger is hit, balancing ordering against holding cost.
Carrying Cost Calculator
Price what buffer and cycle stock cost to hold for a year across storage, capital and shrinkage.
Browse everything on the free tools hub. If receiving and putaway keeps stretching your lead time, see how Locad handles warehousing and storage and multi-market fulfilment.
Reorder Point Questions
Shorter Lead Times Beat Bigger Buffers
Cutting days and variability out of receiving lowers your reorder point without touching service level. Talk to Locad about warehousing and fulfilment for your stock.