Par level: the fixed target, and where it breaks for a growing store
A par level is the fixed quantity of a SKU you want on the shelf at all times: you count what is left, then order enough to bring it back up to that number. It is the simplest inventory method there is, and usually the first one a growing store outgrows.
What a par level is
A par level is the fixed quantity of a SKU you hold at all times: you count what is on the shelf, work out the gap to that number, and order the gap. The term came out of bar and restaurant stock control, where a manager walks the shelves on a set day, writes the shortfall against each line, and sends the order. One number does two jobs at once. It is the signal to act, and it is the target you order up to.
Nothing in it is calculated. A par level is a judgement call: someone who knows the SKU decides sixty is about right and writes sixty on the sheet. That is also its appeal. It takes a minute per SKU, it needs no sales history, and anybody on the team can run the count without learning any arithmetic.
A worked example
An 8 oz sandalwood candle has a par of 60. Monday's count says 38 on the shelf. The gap is 22, so you order 22 and the shelf goes back to 60. Next Monday you count again. No velocity, no lead time, no formula: one fixed number and a repeating count.
Where the fixed number breaks
A par level has no opinion about how fast the SKU is selling or how long the supplier takes. Both of those move, and the par stays where you wrote it.
- Velocity moves. Sixty units is twenty days of cover at three a day, and sixty days of cover at one a day. The same number is over-cautious in a quiet month and far too thin during a promotion.
- Lead time moves. If a supplier quotes fourteen days and slips to twenty-five, a par that held up last quarter now empties the shelf before the pallet lands.
- The count is the trigger. A par level only tells you anything on the day you count. Sell out on a Wednesday and the sheet does not notice until Monday.
- It does not scale by hand. Fifty pars are a morning's work to review. Four hundred are not, so in practice they stop being reviewed, and the store ends up ordering against numbers set a year and a different season ago.
The fix is not a better guess. It is to stop writing a quantity and start deriving one: take the units per day the SKU actually sells, multiply by the days your supplier needs, and let the trigger move when either input moves. That is the reorder point, and we work it through on its own page.
Related terms
Par level sits in a small family of terms that get used interchangeably and should not be.
- Reorder point. The stock level at which you place an order, derived from sales velocity and lead time rather than chosen. A par level is a target quantity; a reorder point is a calculated trigger.
- Min/max levels. The two-number version of the same idea: the min is the trigger, the max is the par. It splits the job a single par level tries to do alone, and it has its own blind spot.
- Safety stock. The buffer you hold on top of expected demand to cover a late delivery or a demand spike. A par level usually has safety stock baked into it invisibly, which is why nobody can say how much of the number is buffer.
- Lead time. The days between sending a purchase order and having sellable stock. The input only you can supply, and the one most often out of date.
- Sales velocity. Units sold per day over a chosen window. Change the window from 7 days to 30 and the same SKU can look like a different product.
- Days of inventory on hand. Current stock divided by daily sales. The number that tells you whether a par of sixty is generous or already late.
Questions
What is the difference between a par level and a reorder point?
A par level is a quantity you choose and top up to. A reorder point is a quantity you calculate from how fast the SKU sells and how long your supplier takes, so it changes when either of those changes. Par levels are set by judgement, reorder points by arithmetic.
How do I set a par level for a SKU?
Most merchants pick a number that has felt safe in the past: enough to cover the gap between counts plus a cushion for a late delivery. If you want to be less arbitrary, work out how many units the SKU sells per day, multiply by the days your supplier needs, and treat that as the floor your par has to clear.
Is a par level the same as min/max inventory?
Not quite. Min/max uses two numbers, a minimum that triggers the order and a maximum you order up to. A par level collapses both into one, which is simpler to run and gives you no room between the trigger and the target.
How often should par levels be reviewed?
As often as the things underneath them move. If a SKU's daily sales shift with the season, or a supplier changes its quoted lead time, the par is wrong until someone edits it. That review cost is the real reason fixed pars drift out of date on a catalogue of a few hundred SKUs.
Can I keep par levels in a spreadsheet?
Yes, and plenty of stores do. The spreadsheet holds the numbers fine; what it cannot do is watch your stock between counts or notice that a SKU's velocity doubled last week. You are the trigger, which works until the week you are busy.
Is a par level ever the right choice?
For a short, stable catalogue counted on a fixed day, a par level is often enough, and it is far better than nothing. The moment velocity varies by season, lead times vary by supplier, or the SKU count passes what one person can review, the fixed number starts costing you either cash tied up in stock or sales you could not fill.