A restaurant can be busy and still be carrying too much stock. Ten cases of cups may feel safe until they tie up cash, crowd the storeroom, or turn out to be the wrong size for the drinks you now sell.
Restaurant par levels give the team a practical target: the quantity you want on hand after ordering, receiving, or finishing prep. They replace "order what we usually order" with a number that has a reason behind it.
Par level versus reorder point
These terms get mixed together, but they answer different questions.
- A reorder point is the quantity that tells you it is time to order, recount, or make a decision.
- A par level is the quantity you want to return to after that decision.
For chicken breast, 8 kg might be the point where a chef checks the next delivery. A par of 24 kg might cover the expected use between deliveries plus a modest buffer. The minimum tells you when to act; the par tells you how far to replenish.
Start with real usage, not shelf capacity
You do not need a complicated spreadsheet to make a defensible first par. Ask four questions for each important item:
- How much do we normally use between dependable deliveries?
- How long does the supplier take to deliver?
- Is the item perishable, expensive, or awkward to store?
- What happens to service if we run out?
A café using 12 litres of milk a day with deliveries every two days might start with a 30-litre par: 24 litres of expected use and a small buffer for a busier morning or late truck. The useful starting formula is:
Typical daily use × days until the next dependable delivery + a reasoned buffer = first par.
The buffer should reflect a real risk: a weekend, supplier cutoff, variable lunch rush, or delivery delay. "We always order extra" is a habit, not an explanation.
Treat items differently
A single rule such as "two weeks of everything" usually creates waste. High-cost proteins need tighter counts and tighter buffers. Fast-moving perishables benefit from shorter delivery cycles. Stable dry goods can carry a larger par if storage is inexpensive. Low-cost but service-critical items--receipt rolls, takeaway lids, cleaning supplies--often deserve a practical backup because a stockout stops work.
Use repeatable units: kilograms, bottles, bags, trays, cases, or portions. The restaurant inventory spreadsheet can give the team a stable list of items and units. If a count later looks wrong, work through the inventory variance guide before jumping to conclusions.
Review pars when the operation changes
Pars are working assumptions, not permanent settings. Revisit them when a supplier changes delivery days, a menu item takes off, a recipe yield changes, storage becomes tighter, or an item is repeatedly discarded. Do not reset a par after one unusually busy Friday; look for a pattern over several cycles.
When the supplier invoice changes, compare cost per usable unit rather than reacting to a higher case total. Our supplier price comparison guide shows how to do that.
A worked par-level example
Consider a small grill that orders chicken three times a week. Over four comparable delivery cycles, the chef sees average use of 10 kg per day. The next reliable delivery arrives in two days. Chicken is a high-cost product, and the walk-in has room for only a little extra without crowding other ingredients.
The first target could be 24 kg: 20 kg for two normal days plus a 4 kg buffer. The reorder point might be 8 kg. That creates a clear instruction. When the count reaches 8 kg, check the delivery schedule and place the order needed to return to 24 kg.
Then test it. If the restaurant repeatedly has 14 kg left before delivery, 24 kg may be too high. If it repeatedly reaches 3 kg on busy Saturdays, the problem may be a weekend delivery gap rather than a universally low par. The answer could be a Friday top-up, not more chicken every day of the week.
This is the important discipline: change one assumption at a time. A par is easier to improve when you can tell whether the issue was usage, delivery timing, buffer, or the count itself.
How to count toward a par without wasting time
A full inventory count has its place, but it is not the only useful count. For par work, create a short "must-count" list for each order day. Include the products that are expensive, regularly run low, or have caused service problems. Count them at a predictable time, in the same unit, before the person responsible for ordering begins.
For each item, write four columns:
| Item | Usable on hand | Par | Order amount |
|---|---|---|---|
| Chicken breast | 11 kg | 24 kg | 13 kg |
| Whole milk | 14 L | 30 L | 16 L |
| Burger buns | 38 | 120 | 82 |
The order amount is simply par minus usable stock. "Usable" matters. Do not treat a leaking pack, expired dairy, product allocated to a catering job, or an unopened case in the wrong unit as if it is freely available for normal service.
A count should also include a brief exception note. If the milk count is low because a delivery was delayed, say that. If chicken is high because a weekend event was cancelled, say that. Those notes stop a sensible temporary decision from becoming a mysterious permanent change next week.
Common par-level mistakes
The most common mistake is setting every par from the last emergency. A restaurant runs out of limes one Friday, raises the lime par dramatically, then throws them away two weeks later. Instead, investigate why the stockout happened. Was there a booking? Was the count wrong? Was the supplier late? Did a drink promotion change usage?
Another mistake is ignoring yield. A 10 kg case of produce does not always provide 10 kg of usable product. If trimming or spoilage is normal, set the par around the usable amount, and make that yield visible in the ordering method.
The third mistake is treating the par list as a substitute for communication. A number cannot tell the next manager that a supplier is out of stock, a freezer is unreliable, or an item was moved for an event. Pair the list with a short handover note when something changes.
A monthly par review that takes 20 minutes
Once the first numbers are in place, schedule a quick monthly review. Pull the items with the most stockouts, the most waste, and the biggest value on hand. For each one, ask: did we use the right unit? Did we order to the stated par? Did the delivery happen when expected? Did the buffer cover a predictable risk, or did it just add cost?
You are not trying to make every item mathematically perfect. You are trying to reduce repeated surprises. A restaurant with sensible pars still needs judgment during holidays, weather events, menu changes, and supplier disruption. The list gives that judgment a consistent starting point.
Where TableAI fits
TableAI can preserve stock updates sent through WhatsApp and compare a reported quantity with a minimum you set. It does not forecast demand, create purchase orders, or decide pars for you. The useful work remains in the restaurant: count honestly, choose a sensible unit, and change the target when the operation changes.
Common questions
Are par levels the same as a budget? No. A par is a quantity target; a budget is money. They affect each other, but they are not interchangeable.
Should every item have a par? Start with items that are expensive, service-critical, or easy to run out of. Expand only when the routine is reliable.
What if people keep ordering above par? Ask why. The par may be unrealistic, the delivery may be unreliable, or the count may be poor. Fix the cause instead of treating the sheet as a policing tool.