Inventory management for a Shopify subscription box business

You are not managing inventory in general. You are counting one thing: whether every active subscriber gets the box you promised, with the SKUs you already locked, on the day the renewal run fires.

The work usually looks like a spreadsheet tab per cycle. Subscriber count times units per box. Then a guess for churn, a guess for new signups, a line for breakage, and a column for whatever the one-off retail side of the store will eat out of the same stock pool before renewal day. You lock curation, you email the supplier, and then you wait out a lead time you do not control.

Two things go wrong, and neither is exotic. The first is that retail quietly drains a SKU you had earmarked for the box, so the count you made three weeks ago is no longer true. The second is that one small component, a sample sachet, a card, a 30ml tester, falls short, and a shortage of one line holds up every box you were going to pack. Running out is a real financial hit, and on a box model it hits twice: the margin on the units you cannot ship, and the subscriber who gets a swap they did not ask for.

A common ask in this category is modest: not a full planning suite, just a clear signal of when to reorder. That is the job Restoc is made for. It watches stock on your store and raises a hand when a SKU crosses its reorder point, so the decision reaches you while there is still time to place an order.

A subscription box has the lumpiest demand curve in commerce. Most of a cycle's units leave in a day or two, then the shelves sit still. Any average taken over too short a stretch will tell you something untrue: measured across renewal day, a SKU looks like it is flying; measured the week after, it looks dead.

Restoc calculates a reorder point per SKU from three inputs you control. A sales velocity window, set to 7, 14 or 30 days. The lead time that applies to that SKU. And a store-wide safety stock percentage that sits on top of lead-time demand.

Picking the window

For a monthly cadence, a 30-day window is usually the honest one: it holds a whole cycle, so the spike and the quiet weeks average out instead of fighting each other. A 7-day window suits a store whose sales arrive evenly, which a box brand's rarely do. The window is set store-wide, so you pick the one that fits the bulk of your catalogue. On Pro you can override the window for an individual SKU, which is worth having if a few retail fast movers behave nothing like your box components.

Lead time, by vendor or by SKU

A reorder point is only useful if it fires early enough to be acted on. A component with a six-week supplier needs its alert six weeks before the shelf empties, not when the shelf looks low. Restoc does not guess your lead times, and it does not pull them from anywhere. You set them, in whichever of two ways costs you less time.

Per vendor is the quick route. The scheduled sync reads the vendor Shopify records on each product, which for most box brands is the supplier under another name, and a vendor lead times screen in settings lists every distinct vendor in your catalogue. Set one number against a vendor and every SKU under it that has no lead time of its own inherits it. You are not typing a figure into hundreds of rows one at a time, and when a supplier slips from four weeks to seven you change it once.

Per SKU is the exception route, for the lines you want to be exact about. A SKU with its own lead time keeps it, whatever its vendor says.

Safety stock for the lines that sell in bursts

An average-based reorder point assumes demand arrives evenly. A box brand's does not, so a line can run out before the average ever reaches the trigger. Restoc takes a store-wide safety stock percentage and adds it on top of lead-time demand, and then shows the resulting cushion in units: on the dashboard, in the weekly report and in the CSV export. You can see what the number is carrying rather than trusting arithmetic you cannot inspect.