Capacity is three multiplications and one of them gets forgotten
Facings across, units behind each facing, layers high. People count the first and assume the second, and the second is where most of the capacity actually is. At the defaults a 36-inch block of a 3.25-inch product holds ten facings, but the deck is 16 inches deep and the product is four inches deep, so there are four units behind every face. The block holds forty, not ten.
That matters because the deck depth is usually free and the deck width is not. Widening a block takes space off a neighbour. Using the depth you already have takes nothing off anybody — it costs only the willingness to load a block properly rather than facing up a single row and leaving the back of the deck empty, which is what a shelf looks like when nobody has time.
Why the trip count is the number that matters
Days of cover is the readable output but trips per week is the one that costs money. At 84 units a week with 35 units of usable stock on the shelf, the block gets refilled 2.4 times a week — 125 times a year, six minutes a trip, about twelve and a half hours. For one line. A shop with 400 lines does not have 400 of those, because most lines are slow, but the twenty fastest lines can easily account for more replenishment labour than the other 380 together.
The shape of the relationship is the useful part. Trips go as one over the facing count, so the saving from each added facing shrinks. Going from four facings to five on this item cuts trips from about 7.6 a week to about 5.6 — two whole trips. Going from ten to eleven cuts them from 2.40 to 2.15, a quarter of a trip. Same one facing, an eighth of the benefit. If you are choosing which blocks to widen, the thin ones on fast lines are where the labour is, and it is not close.
The refill trigger is not zero
Nobody waits for an empty peg. The block gets filled when it looks thin, and that judgement is worth making explicit because it eats capacity you paid for. A trigger of five units on a forty-unit block means you are cycling 35 units, not 40, and the trips go up by about an eighth. Set it to fifteen and you are cycling 25, and trips go up by more than half.
There is a real argument for a high trigger — a block that hits genuine zero has stopped selling and you cannot tell from the till whether anybody tried to buy it — and a real argument for a low one, which is the labour. What there is no argument for is having the number be accidental, which is what it is when nobody has ever said it out loud.
What this does not model
It assumes even selling and one refill event, and shelves do neither. A weekend is not a seventh of the week. A promotion is not a normal week at all. One customer clearing the block for a party takes out two days of cover in a minute and the model has no way to see it coming. Everything here is the steady-state answer, which is the right starting point and the wrong finishing point.
It also says nothing about whether the item deserves the deck in the first place. That question is settled by what the space earns, not by what it holds, and the sales per square foot calculator is where the comparison between categories gets made. Getting a slow line down to one refill a fortnight is not an achievement if the block should have gone to something else.
Related pages
The deck you are dividing up comes from the store fixture run layout calculator, which turns a footprint into linear feet of face. What sits behind the shelf is the stockroom and delivery frequency calculator, and how fast the whole stockholding cycles is inventory turnover. For the order quantity itself, the inventory order calculator covers reorder points and safety stock.
For the pure geometry of fitting objects on a shelf without the retail framing, the pantry shelf capacity calculator does the same packing arithmetic for a household, and the shelf fill and shifting calculator handles the library version, where the problem is not refilling but the cost of inserting into a full run.
Questions people ask
How many facings should a product get?
The version of the question this page can answer is: how many facings does it take before refilling stops being a daily job. Work out the units a block holds at your product size, divide by the weekly sell rate, and read the trips column. Most people set a target of one refill per delivery cycle and let the facing count fall out of that. What no calculator can settle is the other half — whether the deck is better spent on something else — because that is a comparison of what each line earns per foot, not of what it holds.
Does adding facings really reduce restocking work?
Yes, but with sharply diminishing returns, and the shape is worth internalising. Trips per week go as one over the facing count. Doubling a two-facing block halves the trips; adding one facing to a twelve-facing block removes about eight percent of them. So the labour argument for wide blocks is strong on thin, fast-moving lines and almost meaningless on lines that already have plenty of deck. If you are looking for replenishment hours to remove, the thin blocks on your fastest twenty lines are where they are.
How deep should I load a shelf?
As deep as somebody can reliably reach and rotate, which is a measurement of your deck and your staff rather than a rule. The input here is usable depth for exactly that reason: a 20-inch deck where the back four inches never get emptied is a 16-inch deck, and pretending otherwise gives you a capacity figure that is wrong and stock that ages out at the back. If units behind the face cannot be rotated, the honest thing is to reduce the usable depth rather than to accept a capacity you will not actually cycle.
What does days of cover mean on a shelf?
How long the units on the shelf last at the current sell rate, with no refill. It is the shelf-level version of the days-of-supply figure that gets calculated for a whole business, and it is far shorter than people expect — usually two to five days for a moving line, which is why replenishment is a daily activity in most shops rather than a weekly one. The figure on this page is a steady-state estimate from an even sell rate, so treat it as the best case for a normal week.
Should I count the stockroom units as cover?
For deciding when to reorder, yes — that is total on-hand cover and it is the right input to an order decision. For deciding facings, no, because stock in the back does not stop a shelf looking empty and does not save the trip that fills it. The two numbers answer different questions, which is why this page shows them separately. A block with two days of shelf cover and three weeks in the back still gets walked to twice a week.