How to Calculate Reorder Point, Safety Stock and Par
Merchant Education & Growth

How to Calculate Reorder Point and Par Levels Without Overbuying

Learn how to calculate reorder point, safety stock and par levels with worked examples, then set POS low-stock alerts so you stop running out of top sellers.

Gloved hand holding a handheld barcode scanner in front of stacked cardboard boxes in a warehouse

Photo: Pixabay

On this page 7 sections
  1. Reorder Point or Par Level: Pick by How You Order
  2. How to Calculate a Reorder Point in Four Steps
  3. Two Safety Stock Formulas, Two Very Different Buffers
  4. Weekly Order Days Need a Bigger Buffer
  5. Setting Par Level Inventory for Restaurant Ingredients
  6. Turning Reorder Points Into POS Low-Stock Alerts
  7. Recalculate Before the Season Changes

Key takeaways

  • Work out average daily sales using only the days the item was in stock. If you count sold-out days as zero, your average comes out too low and you'll run out again.
  • A 95% service level means a 5% chance of running out in each order cycle. An item you reorder often will still run out a few times a year.
  • If you can only order on a set day, size both expected demand and safety stock to the days between orders plus lead time, not to lead time alone.

Here's how to calculate reorder point for any item you can order whenever you need it: multiply its average daily sales by your supplier's lead time in days, then add safety stock. When the amount you have on hand drops to that number, you place the order. A par level answers a different question for businesses that order on a fixed schedule: how much you need after each delivery to last until the next one, plus a buffer.

Both numbers come from three inputs: how fast an item sells, how long the supplier takes, and how much those two vary. You already have the data. Your POS records every sale, and your invoices show when each order actually arrived.

Reorder Point or Par Level: Pick by How You Order

The choice depends on what triggers your orders.

Reorder points fit businesses that can buy on any day an item runs low. A convenience grocer restocks a snack item when it drops to a set count, and a liquor store does the same with a fast-moving vodka. The stock level triggers the order, and the date doesn't matter.

Par levels fit businesses that order on a calendar. A restaurant with produce deliveries on Tuesday and Friday counts the walk-in before each order and buys the difference between what's on hand and par.

Plenty of businesses run both. A bar might keep pars on the beer and spirits it orders weekly from a distributor, and reorder points on the straws, napkins and cleaning supplies it buys online when needed. If you're still deciding what kind of system should hold these numbers, our rundown of the types of POS systems and which fits your business is a good place to start.

How to Calculate a Reorder Point in Four Steps

This is the standard formula, the same one in NetSuite's inventory guide and most planning software:

Reorder point = (average daily sales × lead time in days) + safety stock

  1. Pull average daily sales from your POS. For steady items, two or three months of history smooths out week-to-week swings. Divide units sold by the number of days the item was actually in stock. Many merchants get this wrong. If an item sat sold out for six days and you count those as zero-sales days, your average comes out low and you'll run out again.
  2. Measure lead time from order to sellable. Count from the moment you place the order to the moment the item is on the shelf or prepped, and include the supplier's order cutoff. If the distributor needs orders by 2 p.m. and you usually send yours at 4 p.m., you've added a day. Use the dates on recent invoices, not the number the sales rep quoted.
  3. Add safety stock to cover busy days and late trucks. The next section compares two ways to size it.
  4. Enter the result in your POS as the item's low-stock threshold.

Worked example (illustrative numbers). A liquor store sells a mid-priced bourbon. Over its last 60 in-stock days it sold 240 bottles, an average of 4 a day. The distributor usually delivers 3 days after an order. The store's best day in that stretch sold 6 bottles, and its slowest delivery took 4 days.

Expected sales during the lead time are 4 × 3 = 12 bottles. Using the max-minus-average method explained below, safety stock is (6 × 4) − (4 × 3) = 24 − 12 = 12 bottles. The reorder point is 12 + 12 = 24. When the shelf and backroom together hit 24 bottles, the store orders.

Two Safety Stock Formulas, Two Very Different Buffers

Safety stock covers the gap between an average day and a bad one. The two common methods can give very different answers, so choose one deliberately.

Max Minus Average: Simple and Cautious

Safety stock = (max daily sales × max lead time) − (average daily sales × average lead time)

NetSuite lists this as the basic safety stock formula. It needs only four numbers from your sales reports and invoices. Its weakness is that it assumes your worst sales day and your slowest delivery happen at the same time. That's rare, so the buffer runs high. For the bourbon it called for 12 bottles.

The Statistical Method: Sized to a Target

Safety stock = Z × standard deviation of daily sales × √(lead time in days)

Z is a service-level factor taken from the normal distribution. Published Z tables, such as the one in Ordio's safety stock calculator, give 1.28 for a 90% service level, 1.65 for 95%, 1.88 for 97% and 2.33 for 99%. The standard deviation measures how much daily sales bounce around. To get it, put daily unit sales in one spreadsheet column and use the STDEV.S function in Excel or Google Sheets.

Same bourbon, statistical method (illustrative). Say daily sales have a standard deviation of 1.5 bottles. At a 95% target, safety stock = 1.65 × 1.5 × √3 = 1.65 × 1.5 × 1.73 ≈ 4.3, rounded up to 5 bottles. The reorder point drops from 24 to 17.

Seven bottles of one bourbon isn't much money. Repeat that gap across a few hundred items, though, and it adds up to real cash sitting on shelves.

What a 95% Service Level Actually Promises

This is the most misread number in inventory planning. A 95% service level does not mean 95% of customers find the item. It means a 95% chance of not running out during any one replenishment cycle, which is the window between hitting the reorder point and the delivery arriving.

So the number of stockouts to expect depends on how often you reorder. Say the store above orders two cases (24 bottles) at a time. It reorders about every 6 days, or roughly 60 times a year. A 5% miss rate per cycle works out to about 3 stockouts a year on that one item (illustrative). Most will be short, because they happen at the very end of a cycle, but they will happen. If 3 a year is too many for a particular item, raise Z instead of guessing at a bigger number.

Which Method to Use

Use the statistical method for steady sellers where you have a few months of daily sales. Use max minus average, or a Z of 2.33, for items where running out costs more than the product itself: the house well vodka, the milk a grocery customer came in for, the brisket your menu is built around. The statistical formula also assumes lead times are fairly stable. If a supplier's deliveries swing by days, size the buffer up or switch that item to max minus average.

Weekly Order Days Need a Bigger Buffer

Both formulas assume you order on the day you hit the threshold. Many merchants can't. If your liquor distributor takes orders only on Mondays, a bottle that crosses the reorder point on Tuesday waits six days before you can even place the order.

With a fixed order day, you're running a par system whether you call it that or not. Set an order-up-to level instead: average daily sales × (days between orders + lead time), plus safety stock. The buffer has to cover that whole window too. In the statistical formula, replace lead time with days between orders plus lead time.

For the bourbon, ordered weekly with a 3-day lead time, the window is 10 days. Expected sales are 4 × 10 = 40 bottles. Safety stock is 1.65 × 1.5 × √10 ≈ 7.8, rounded up to 8. The order-up-to level is 48 bottles (illustrative). Each Monday, the store orders the difference between 48 and what's on hand. Keeping the 5-bottle buffer sized for a 3-day lead time would leave the store short of protection for most of the week without anyone noticing.

Setting Par Level Inventory for Restaurant Ingredients

In a kitchen, par level inventory starts with how much you use between deliveries, not with daily sales.

Par = expected usage until the next delivery + safety buffer Order quantity = par − on hand, rounded up to the case or pack size

Delivery gaps are rarely equal, so one par per item usually isn't enough. Set a par for each delivery day.

Worked example (illustrative numbers). A restaurant gets chicken thighs on Tuesday and Friday. It uses about 30 lb a day Tuesday through Thursday. Friday through Monday it uses 50, 60, 45 and 25 lb.

  • Tuesday's delivery covers 3 days: 90 lb of usage plus a 20 lb buffer, for a par of 110 lb.
  • Friday's delivery covers 4 days: 180 lb of usage plus a 30 lb buffer, for a par of 210 lb. The weekend buffer is larger because walk-in traffic is harder to predict.

On Friday morning the cook counts 35 lb on hand, so the order is 210 − 35 = 175 lb. The supplier packs 40 lb cases, so that rounds up to 5 cases, or 200 lb.

Shelf life caps par in a way it doesn't for dry goods. If a par is higher than you can use before quality drops, raising it swaps stockouts for waste. The fix is a third weekly delivery or smaller prep batches. Bars have a smaller version of this problem with open wine, draft kegs and fresh juice. Pars on bottled spirits can run higher because they don't spoil.

Restaurants also have a translation problem. The POS sells a chicken sandwich, but the walk-in holds pounds of chicken. For alerts to fire automatically, the system needs recipe-level inventory that subtracts ingredients as menu items sell. Without it, pars live on a clipboard and the physical count drives the order. Ingredient tracking is one of the key feature differences between restaurant and retail POS systems, so check for it before you buy.

Turning Reorder Points Into POS Low-Stock Alerts

The math only helps if it sets off an alert at the right moment. Menu names differ by POS. In Clover, for example, you set a low stock threshold on each item's details page. Clover's help center says an alert icon then appears next to the stock count once the count is at or below that threshold. Clover's April 2026 API update added low-stock alerts to its Inventory API, so apps built on Clover can let merchants set them up too. Whatever system you use, the setup order is similar:

  1. Turn on item-level stock tracking so each sale lowers the on-hand count.
  2. Enter the reorder point as the low-stock threshold. Start with your top sellers by revenue, not all 3,000 items at once.
  3. Fill in vendor, case pack and cost so an alert can become a purchase order without anyone hunting for a price.
  4. Receive every delivery into the POS. If the bourbon arrives and nobody records it, the system still thinks you're out, and the alerts stop meaning anything.
  5. Sync online and in-store stock. If you also run a Shopify store connected to your payments, both channels need to draw from the same count. Otherwise a web order can sell a bottle the register already sold.

An alert is only as good as the on-hand count behind it. Breakage, theft and ringing up the wrong item all make the POS count drift away from what's on the shelf. Regular cycle counts of your fast-moving items fix the count. Tracking inventory shrinkage and where it comes from tells you which items need a bigger buffer or tighter controls.

If your terminal takes payments but can't track stock, you need different hardware or software. MannyPay sells POS systems for retail and restaurant counters, including Clover, Valor and Dejavoo models. The team can help you work out what fits a grocery market with a high item count or a liquor store ordering from distributors.

Recalculate Before the Season Changes

A threshold that's right in February can leave you sold out in December, so reorder points and pars need to be updated on a schedule.

Recalculate regularly. Monthly suits most stores, and weekly suits perishables and anything with a short season. For seasonal items, calculate from the same weeks last year and adjust for this year's trend. A single average across the whole year comes out too low for the busy weeks and too high for the slow ones.

Plan promotions before you run them. Use the sales bump from past promotions on similar items. If a discount weekend doubled a wine's sales last time, double its average daily sales in the formula for the promotion window and order ahead (illustrative). Distributors and produce suppliers are busiest around holidays, so use a longer lead time for those weeks if your invoices show deliveries slipping.

Write a reset date next to every number you raise. The usual mistake is raising a par for the Fourth of July and leaving it there until September. Schedule the drop the same way you'd schedule a promotion's end.

This week, export 60 days of item sales for your 20 highest-revenue items. Calculate a reorder point or order-up-to level for each and enter them as low-stock thresholds. After one full order cycle, compare the alerts with what actually ran out, and adjust the buffers from there. If your current system can't track stock or send alerts, request a free quote and a MannyPay specialist will respond within one business day.

Sources

  1. Safety Stock: What It Is & How to Calculate (NetSuite)
  2. Free Safety Stock Calculator: Formula & Reorder Point (Ordio)
  3. Manage Stock Count and Low Stock Threshold (Clover Help)
  4. Clover API Updates, April 2026 (Clover Developer Docs)

About the author

Isabella Cruz

Isabella Cruz is Business Growth & Commerce Editor for Beyond the Swipe, writing for small businesses, restaurants, retailers, entertainment venues and e-commerce sellers.

This article is for general educational purposes and is not legal, tax or financial advice. Card network rules and state laws change; confirm current requirements for your business.

Related reading