Your reorder point is the stock level at which you place the next order, and the formula is average daily use × lead time in days, plus safety stock. A hot sauce maker who fills 200 bottles a day, waits 14 days for a bottle delivery, and keeps 1,700 bottles as a cushion should reorder when stock falls to 4,500. The arithmetic is easy. What trips people up is where the three inputs come from, so that's most of this article.
The short version
- Work out average daily use from the last 8 to 12 weeks, in calendar days.
- Use the lead time you actually get, in calendar days, not the one on the supplier's website.
- Size the safety stock from the worst week and the slowest delivery you've really had.
- Compare the reorder point against what's on hand plus what's already on order.
- Redo the numbers every quarter, and before your busy season.
The formula, one piece at a time
Reorder point = (average daily use × lead time) + safety stock
The first half is the stock you'll burn through while you wait for the delivery. If that were the whole formula, the truck would pull in just as you used the last bottle, every time, assuming nothing ever went wrong. The second half is for when something does.
Average daily use
Take what you used over the last 8 to 12 weeks and divide by the number of calendar days. Calendar days, even if you only produce four days a week, because your supplier's lead time runs through weekends too. Mixing "production days" on one side and "calendar days" on the other is the most common way to get a reorder point that's 30% too low.
Our hot sauce maker filled 16,800 bottles in the last 12 weeks. That's 84 days, so 200 a day.
Lead time
Count from the day you place the order to the day the goods are on your shelf and usable. The supplier says "10 business days," which is 14 calendar days. Then check your last few orders. If they took 14, 13, 11 and 18 days, your average is 14 and your worst is 18. Use what happened, and keep the worst one. You'll need it in a minute. (There's more on measuring it in the lead time calculation.)
Safety stock
There are formulas for this with square roots and service levels, and they're the right tool if you run a warehouse with a thousand SKUs. For a small producer there's a simpler method that works, and you can explain it to anyone: prepare for the worst case you've actually seen.
Safety stock = (busiest daily use × longest lead time) − (average daily use × average lead time)
The busiest week in the last year was 1,750 bottles, or 250 a day. The slowest delivery took 18 days.
- Worst case: 250 × 18 = 4,500 bottles
- Normal case: 200 × 14 = 2,800 bottles
- Safety stock: 4,500 − 2,800 = 1,700 bottles
So the reorder point is 2,800 + 1,700 = 4,500 bottles. You'll notice that's the same as the worst case. That's the idea: you reorder early enough that even your busiest stretch and your slowest delivery, arriving together, leave you at zero and not below it.
What it looks like over a month
Say the supplier's minimum order is 5,000 bottles.
| Day | What happens | Bottles on hand |
|---|---|---|
| 0 | Stock hits 4,500. You order 5,000 | 4,500 |
| 14 | Normal delivery arrives. You've used 14 × 200 = 2,800 | 1,700 + 5,000 = 6,700 |
| 25 | Stock is back down to 4,500. Order again | 4,500 |
On a normal cycle you never dip below 1,700. That cushion isn't waste. It's what the worst month draws on:
| Day | Busy stretch, slow truck | Bottles on hand |
|---|---|---|
| 0 | Stock hits 4,500. You order 5,000 | 4,500 |
| 14 | No delivery yet. You've used 14 × 250 = 3,500 | 1,000 |
| 18 | Delivery arrives. You've used 18 × 250 = 4,500 | 0 + 5,000 = 5,000 |
You got to zero on the day the truck arrived. Tight, and you kept filling bottles the whole time.
Three items, side by side
Different items need very different reorder points even in the same business, because lead times differ more than usage does.
| Item | Avg use / day | Avg lead time | Busiest use / day | Longest lead time | Lead-time demand | Safety stock | Reorder point |
|---|---|---|---|---|---|---|---|
| 5 oz bottles | 200 | 14 days | 250 | 18 days | 2,800 | 1,700 | 4,500 |
| Front labels | 200 | 21 days | 250 | 28 days | 4,200 | 2,800 | 7,000 |
| Habanero mash | 12 kg | 7 days | 15 kg | 10 days | 84 kg | 66 kg | 150 kg |
Look at the labels. Same usage as the bottles, but the printer takes three weeks and once took four, so the reorder point is 7,000. Labels are the item small producers run out of most, for exactly this reason: they're cheap, nobody thinks about them, and the lead time is the longest in the building.
In a spreadsheet, with average use in B, average lead time in C, busiest use in D and longest lead time in E:
Lead-time demand =B2*C2
Safety stock =(D2*E2)-(B2*C2)
Reorder point =D2*E2
Days of cover now =(on hand + on order) / B2
Yes, with this method the reorder point simplifies to busiest use times longest lead time. I keep the two halves as separate columns anyway, because seeing "1,700 of this is cushion" makes it a lot easier to decide what to do when cash is tight.
The mistakes that cause stockouts anyway
Comparing against on-hand only. If you've got 3,000 bottles on the shelf and 5,000 arriving Thursday, you don't need to order. If you've got 4,400 on the shelf and nothing coming, you're already late. The number to compare is on hand plus on order.
Forgetting the minimum order. If your supplier's minimum is 5,000 and you use 200 a day, every order is 25 days of stock. That's fine, and it means you don't need to check daily. If the minimum is a full pallet that lasts five months, the real question is where you'll store it, which no formula answers.
Setting it once. A reorder point set in February is wrong by November. Recalculate every quarter, and before any season where your sales double.
Using it for things that arrive tomorrow. For butter and cream from a supplier who delivers next day, a reorder point is more math than you need. A par level ("bring it back up to 6 cases every order day") does the job. Par levels and reorder points solve different problems.
Where this stops working
The formula is only as good as the "on hand" number you're comparing against. If the count on your spreadsheet is from two weeks ago, you'll find out you're below the reorder point when someone opens the last case. The other problem is scale. Three items are easy. Eighty items, each with its own usage and lead time, recalculated quarterly, is a job nobody does, so the reorder points quietly go stale while the business changes around them. When the count and the shelf stop agreeing, the reorder math goes with it.
How TaroStack does it
TaroStack keeps the on-hand number current as you receive, make and ship, so there's something true to compare against. It looks at how each item has actually moved and suggests a reorder point for it, which you can accept or change. The reorder list shows every item in days of cover, including what's already on order, with a red line at the point where ordering today would already be too late. It also tracks how long each supplier really takes against what they promised, so the lead time in the math is the real one. When it's time, you can raise one purchase order for everything you need from a supplier in one go.
