Most businesses have a standing answer to this question, and the answer is "about what we ordered last time." It survives because it never gets billed. No invoice ever arrives for ordering the wrong amount, so the habit ages in place, sometimes for years, sometimes from back when the business was half its current size.
The order size has a real answer. Finding it takes three numbers you mostly already know, and checking whether your habit is close takes two.
Two costs, pulling in opposite directions
Every order you place costs something fixed, no matter its size. Someone's twenty minutes on the phone or the portal, a delivery fee, an invoice to process and pay. Order bigger and you order less often, so this cost falls.
Everything you hold costs something too. Shelf and storeroom space, spoilage and shrink, insurance, and the use of the money itself, which is not an abstraction if you carry a balance on a line of credit to own stock that sits there. Order bigger and more sits there longer, so this cost rises.
One cost falls as orders get bigger, the other climbs, and their sum bottoms out at a particular quantity. Consultants call that quantity the economic order quantity, and the formula for it has been settled since 1913. You can use the idea without ever touching the formula. What matters is knowing the bottom exists, roughly where it is, and which side of it your current habit sits on.
A worked example, in tubes of hair color
Take a hypothetical salon that goes through 40 tubes of color a week, about 2,080 a year. Placing an order costs it roughly $15 once the phone call, the delivery minimum, and the invoice handling are counted honestly. Carrying a tube for a year costs about $3, using the rough rule that holding stock runs 20% to 30% of what the item cost you.
Run those three numbers and the answer comes back. Order about 144 tubes at a time, call it three and a half weeks of supply, fourteen or so orders a year.
Here is the part worth noticing. The owner of this salon currently orders weekly, 40 tubes at a time, because weekly feels careful. Careful is costing her money on both ends. Fifty-two orders a year at $15 is $780 of ordering cost, against $60 of carrying cost for the small stock she keeps. The 144-tube habit spends $216 on each. Total, about $432 a year against her $840. Not life-changing on one supply line. Multiply it across everything the business orders, and it usually is a real number, recovered without changing anything except when the phone gets picked up.
The check hiding in the answer
At the right order size, something tidy happens. Your annual ordering cost and your annual carrying cost come out equal. In the example, $216 and $216. That is not a coincidence. It falls out of the arithmetic every time, and it hands you a diagnostic you can run without any calculator at all.
Add up what a year of placing orders costs you for one supplier. Add up what a year of holding that supplier's stock costs you. If the two are wildly unequal, your order size is wrong, and the bigger side tells you the direction. Ordering cost dwarfing carrying cost means you are ordering too little, too often, like the salon. Carrying cost dwarfing ordering cost means the reverse, and your cash is sitting on a shelf as boxes. Two numbers, one comparison, and you know which way to move before you ever touch a formula.
You can be sloppy about it, and should be
The formula's answer is 144, and your distributor sells cases of 24. Order six cases. The curve is remarkably flat near its bottom, which means being roughly right is cheap. Miss the ideal quantity by 20% in either direction and your total cost rises by less than 2%. The habit that hurts is sitting at double the right quantity, or at half of it, which costs about 25% extra and is exactly where "what we ordered last time" tends to drift while a business grows around it.
Growth bends the answer in a way worth knowing about. The right order size scales with the square root of demand, so a business that doubles should order about 1.4 times as much per order, and order 1.4 times as often. An owner who doubles the order because sales doubled has overshot. The habit that was right at one size is quietly wrong at the next one, in a knowable direction, by a knowable amount.
What the answer leaves out
The math assumes the price per unit stays the same at any order size, and suppliers do not always cooperate. A genuine volume discount changes the calculation, sometimes decisively, and the honest move is to run the total cost both ways rather than assume the discount wins. It often does. It is not free just because the per-unit price fell, since the extra stock still costs you money to hold.
Perishables put a ceiling on order size that has nothing to do with this arithmetic. And a business with a real season should run the numbers separately for the busy stretch and the slow one rather than average across the year, for the same reason the reorder question needs seasonal treatment, because an average of two different businesses describes neither.
The carrying cost is the number most owners have never actually worked out, and the 20-to-30% rule is a hand-wave standing in for it. It deserves better than a hand-wave, because everything above moves when it moves. That one is its own conversation.
If you want the arithmetic done for you, the calculator runs it in your browser from your own numbers, alongside the when-to-reorder half of the problem. Nothing you type is saved or sent anywhere. Or start smaller than that. Pull one supplier's invoices and run the two-number check. The direction it points is the whole diagnosis, and you probably already suspect which way it goes.