🔒 100% private — runs in your browser, nothing is uploaded

Reorder Point & Safety Stock Calculator

From average daily demand, its spread, lead time and a target service level: the safety stock to hold and the stock level that should trigger the next order.

A reorder point answers a timing question: at what stock level do you place the next order so the shelf does not empty while the delivery is on its way? Two numbers build it. The demand you expect during the lead time — average daily demand multiplied by the lead time in days — covers the normal case. The safety stock covers the abnormal one: demand running hotter than average, or the truck arriving late. Run out before the delivery and the model has failed at exactly the job it exists for, so the safety stock is sized from how much demand varies and how often you are willing to risk a stockout — your service level.

Two numbers decide when to reorder: the demand you expect while the delivery is on its way, and a safety stock sized from how much that demand (and the delivery itself) tends to vary — and from how often you can afford to run out. The formulas are printed in full below the tool, and the defaults are a worked example you can check by hand.

Planning aid only — not business or financial advice. The model assumes roughly normal, independent daily demand; promotions, seasonality and lumpy demand break it, and the page says so rather than hide it. Pair this with a winning quote from the Supplier Quote Comparison and the chain tools on the Seller Procurement Workbench.

Data passport — where the figures on this page come from

Data sourceThe standard safety-stock model (operations management references; APICS/ASCM body of knowledge) with one-sided standard-normal z factors for the service levels
EffectiveCurrent published model — unchanged
Last verifiedNovember 2026
Watch ownerSeller-cluster data watch — annual re-check against the published model
⚠️ Estimate only — this calculator is a planning aid, not business, operations or financial advice. It applies the standard textbook safety-stock model, which assumes demand is roughly normally distributed, daily demands are independent of each other, and the lead time and demand figures you enter describe the future the way they described the past. Promotions, seasonality, trends, supplier slowdowns and lumpy (intermittent) demand all break those assumptions; for slow-moving or sporadic items this model over- or under-stocks and specialist methods exist. The service level is the probability of not stocking out during one replenishment cycle under the model — not a promise about your fill rate. Last verified: November 2026.

Quick answers

How do you calculate a reorder point?

Reorder point = demand during lead time + safety stock. Demand during lead time is average daily demand × lead time in days. Safety stock is z × demand standard deviation × √ lead time, where z comes from your target service level (95% → 1.645). Worked: 50 units a day, σ = 12, 7-day lead time, 95% → safety stock 52.23, lead-time demand 350, reorder point 402.23 — order at 403 units.

Round the result up, never down: the fractional part is part of the cover you sized. If your lead time itself varies, use the combined formula below the tool — it adds a second term for lead-time spread and the safety stock grows, correctly, because late deliveries are the other half of stockout risk.

How much safety stock do I need at a 95% service level?

Safety stock = 1.645 × σ<sub>daily</sub> × √ lead time. With daily demand averaging 50 and a standard deviation of 12 over a 7-day lead time, that is 1.645 × 12 × 2.6458 = 52.23 units — hold 53. At 99% (z = 2.326) the same figures need 73.85 units: the last few points of service level are the expensive ones.

The service level is the probability, under the model, of not stocking out during a single replenishment cycle. It is not the share of orders filled (that is the fill rate, a different metric that usually comes out higher). Choose the level per item: a stockout on your best seller costs more than one on a slow line, and the z table under the tool prices each step.

What changes when the lead time itself is unreliable?

Use the combined spread: σ over the lead time = √(lead time × σ<sub>demand</sub>² + average demand² × σ<sub>lead</sub>²), then safety stock = z × that. With demand 50 ± 12 a day, a 7-day lead time that itself varies by σ = 1.5 days, at 95%: σ = √(1008 + 5625) = 81.44, safety stock 133.97 instead of 52.23 — an unreliable lead time more than doubles the buffer here.

That jump is the honest arithmetic of late deliveries: a day of delay at full demand costs a whole day’s sales of cover, so lead-time spread is weighted by demand itself. If the combined figure shocks you, the lever is the supplier relationship (or a second source), not a smaller z chosen to feel better.

How to use

  1. Enter average demand — either directly as units per day, or as units sold over a period with the period’s length in days, and the page divides.
  2. Enter the standard deviation of daily demand (from your sales history — a spreadsheet’s STDEV over daily unit sales is exactly this figure).
  3. Enter the supplier lead time in days. If it varies, tick the box and add its standard deviation too — a quoted "5–10 days" is roughly a mean of 7.5 with a spread you can estimate from past deliveries.
  4. Pick the target service level (or enter your own z). The safety stock, the reorder point and the full one-line working update as you type.
  5. Take the rounded-up reorder point into whatever system actually tracks your stock. This page stores nothing — the numbers are yours to carry.

About this tool

The service level enters through the standard normal factor z: 90% gives z = 1.282, 95% gives 1.645, 99% gives 2.326. Higher service costs disproportionately more stock — moving from 95% to 99% raises the safety stock by about 41% on the same demand spread, which is why the level is a decision, not a default. The calculator shows its working in one line — the spread of demand over the lead time, the multiplication by z, and the final addition — so any figure it gives can be checked by hand against the formula printed on this page.

This page deliberately stops at the calculation. It holds no inventory, watches no stock levels and sends no reminders: continuous stock management needs your live sales and stock data on a server, and this site’s seller tools are built to keep business data in your browser instead. You bring the demand figures from your own records, take the two numbers away, and use them wherever you actually manage stock.

How it works

Reorder point = lead-time demand + safety stock. Lead-time demand = average daily demand (d) × lead time in days (L).

Safety stock, constant lead time = z × σ<sub>d</sub> × √L, where σ<sub>d</sub> is the standard deviation of daily demand and z is the standard-normal factor for the target service level (90% → 1.282, 95% → 1.645, 99% → 2.326 — full table below). Demand over L days varies less than L separate worst days added together, which is what the square root expresses.

Safety stock, variable lead time = z × √(L × σ<sub>d</sub>² + d² × σ<sub>L</sub>²). The first term is demand spread accumulated over the lead time; the second is lead-time spread weighted by demand. With σ<sub>L</sub> = 0 the formula reduces to the constant-lead-time one — the page switches on the values you give it, and the working line under the result shows which form was used.

Worked examples

The default case — steady demand, fixed lead time (check it by hand): Average daily demand 50 units, standard deviation 12, lead time 7 days, service level 95% (z = 1.645). Spread over the lead time: 12 × √7 = 12 × 2.6458 = 31.7490. Safety stock: 1.645 × 31.7490 = 52.23 units — hold 53. Lead-time demand: 50 × 7 = 350. Reorder point: 350 + 52.23 = 402.23 — place the next order when stock falls to 403 units. These are the calculator’s default inputs, so the live page shows exactly this result.

Same item, unreliable lead time: Same demand (50 ± 12 a day, 95% service), but the 7-day lead time varies with a standard deviation of 1.5 days. Combined spread: √(7 × 144 + 2500 × 2.25) = √(1008 + 5625) = √6633 = 81.4432. Safety stock: 1.645 × 81.4432 = 133.97 — hold 134. Reorder point: 350 + 133.97 = 483.97 — order at 484 units. Tick the lead-time variability box on the defaults to watch the buffer grow.

Working daily demand out from a month of sales: You sold 1,500 units over the last 30 days: choose "sales over a period", enter 1,500 units and 30 days, and the page uses 50 units a day — the same default case as the first example. Use a period that resembles the season you are ordering into; last December’s daily rate is a poor guide for March.

Service level and the z factor — what each step costs

Target service levelz factorSafety stock on the worked example (σLT = 31.7490)
80%0.84226.73 units
85%1.03632.89 units
90%1.28240.70 units
95%1.64552.23 units
97.5%1.96062.23 units
99%2.32673.85 units
99.9%3.09098.10 units

z factors are the standard one-sided values of the normal distribution (the service level is a cycle service level — the probability of no stockout during one replenishment cycle, under the model). Sources: the standard safety-stock model as set out in operations-management references (e.g. Krajewski et al., Operations Management; APICS/ASCM body of knowledge). Data passport: no external data feed — the model and the z table are the page’s own reference content, last verified November 2026, on the seller-cluster annual re-check.

Frequently asked questions

Where do I get the standard deviation of daily demand?

From your own sales history: list the units sold per day over a representative stretch (30–90 days), and take the standard deviation — STDEV in a spreadsheet — of that column. Use the same days you took the average from. If your history is a weekly total, divide the weekly standard deviation by √7 to approximate the daily figure rather than typing the weekly one in directly.

What service level should I choose?

It prices the stockout risk you accept per replenishment cycle, under the model: 90% accepts about one stockout cycle in ten, 99% about one in a hundred — at roughly 1.8× the safety stock of 90% on the same demand. Best sellers and items with long, unreliable lead times usually earn 95–99%; slow, cheap or easily substituted lines often sit at 85–90%. There is no universal right answer, which is why the z table is printed on the page with the cost of each step visible.

Does this calculator track my stock and remind me to order?

No — deliberately. Tracking needs your live stock and sales data held somewhere between visits, and this site’s rule for business data is that it stays in your browser, in memory, for one visit. The calculator gives you the two numbers; your inventory system, spreadsheet or marketplace dashboard — the place that already holds the data — is where the trigger belongs.

My item sells a few units a week, in bursts. Will this work?

Poorly, and the page would rather say so: the normal-distribution model assumes demand arrives in a fairly smooth flow. For intermittent demand (many zero days, occasional bursts) it mis-sizes the buffer both ways. Methods built for that pattern exist (Croston’s method and its variants); use one of those, or hold a judgement-based buffer and review it often.

Should I round the reorder point up or down?

Up. The formulas give exact cover for the service level you chose; rounding down silently chooses a lower service level than the one you picked. The page shows the exact figure for checking and the rounded-up figure for use.

Why does a longer lead time not simply multiply the safety stock?

Because daily ups and downs partly cancel: over 7 days the spread grows with √7 (≈ 2.65), not 7. That is the square root in the formula, and it is also why safety stock per day of lead time falls as lead times grow — while total cover still rises, through the lead-time demand term, which does scale linearly.

📦 Embed this tool on your site

<iframe src="https://verystep.com/embed/reorder-point-calculator/" title="Reorder Point & Safety Stock Calculator" width="100%" height="520" loading="lazy" style="border:0"></iframe>

Free to use. The embed shows a small "Powered by Verystep" link.

First time embedding? Read the step-by-step guide →

More free tools